Click here to watch a video clip showing colorectal endoscopic submucosal dissection

For many years, conventional endoscopic mucosal resection (EMR) and surgery were the only options for treating a large (>20 mm) sessile or flat colorectal lesion. Conventional EMR, however, often results in piecemeal removal and there is a significant local recurrence rate ranging from 7.4% to 17%.1 2 3 Full histological evaluation is also difficult.

Endoscopic submucosal dissection (ESD), pioneered in Japan for treating early upper gastro-intestinal malignancy, was introduced in the late 1990s by Yamamoto et al4 and Fujishiro et al5 to treat colorectal lesions. The technique has an advantage over EMR in that its effectiveness is not limited by size or shape of the lesion. In the past decade, colorectal ESD has been shown to be superior to EMR, in terms of higher en-bloc resection rate and lower recurrence rate.6 Colorectal ESD can be applied not only to adenoma, but also to intramucosal carcinoma and low-risk submucosal carcinoma, as defined by the Paris classification7 and the Japanese Society for Cancer of the Colon and Rectum.8 Recently, a large-scale multicentre study has shown that ESD alone is adequate in the management of patients with low-risk submucosal carcinoma and achieves an excellent 5-year recurrence-free survival of 98% and 5-year overall survival of 94%.9

Despite the growing evidence to support the use of colorectal ESD, it is not established as a standard treatment outside Japan. The drawbacks of colorectal ESD include longer operating time6 and higher complication rates, especially perforation. Although the perforation rate of ESD is much higher than that of EMR, most ESD perforations can be treated conservatively by clip closure during endoscopy.10 11 In a multicentre study of iatrogenic perforations in Japan, the respective EMR and ESD perforation rate was 0.58% and 14% in 15 160 therapeutic colonoscopies.12 Endoscopic clipping failed in 43.5% of ESD perforations and surgical intervention was necessary.12

Perhaps one of the major hurdles to its general application is that it is a technically demanding procedure that is difficult to set up at a low-volume district hospital. We would like to share our experience of applying this novel technique in a district hospital in Hong Kong.

North District Hospital was a district hospital serving 700 000 population with a case volume of 15 to 20 cases of ESD per year. Since the introduction of the ESD technique at the hospital in 2009, all lateral spreading tumours larger than 2 cm or those unable to be resected en bloc by conventional polypectomy were referred to a single endoscopist to determine the appropriateness of ESD. Colonoscopy was repeated by a single endoscopist to determine the location, size, and nature of each tumour by white-light and narrow band imaging (NBI). Benign polyps not amendable to removal by EMR were triaged to ESD. Target biopsy was performed on Sano III lesions that were triaged to conventional laparoscopic colectomy. No tumours were excluded based on location.

Evaluation of the depth of invasion is important to determine the treatment strategy. To predict the depth of invasion, we used NBI colonoscopy, based on Sano’s capillary pattern classification. The underlying principle is that angiogenesis is critical for transition of a premalignant lesion to a malignant one and the microcapillary pattern changes in this process. Sano et al13 focused on this microcapillary difference based on their histopathological findings and devised three classifications: types I, II, and III. Type III was further subdivided into IIIA and IIIB.14

The diagnostic accuracy of NBI colonoscopy in differentiating a neoplastic from non-neoplastic lesion is superior to conventional colonoscopy and equivalent to chromoendoscopy using indigocarmine.15 For estimation of the depth of invasion, the sensitivity, specificity, and diagnostic accuracy of capillary pattern type III for differentiating pM-ca (intramucosal) or pSM1 (superficial) from pSM2-3 (deep) was 84.8%, 88.7% and 87.7%, respectively.14 We preferred NBI colonoscopy because it is fast and easy to use, without the need to spray dye as in chromoendoscopy.

The procedure was performed in the operating theatre with the patient under conscious sedation. All patients were assessed by an anaesthetist in a preoperation clinic.

Patients were instructed to eat a low-residue diet 2 days before the procedure and a fluid diet on the day before ESD. Bowel preparation with 4 L polyethylene glycol solution was given on the day before ESD. Prophylactic antibiotics were not prescribed.

All procedures were performed in the operating theatre. This ensured that all equipment was on hand should conversion to an open procedure be required, for example, if there was full-thickness perforation that could not be closed endoscopically.

In our hospital, ESD was performed using a single-channel colonoscope (CF-H180AL; Olympus, Tokyo, Japan). This colonoscope was a high-density television compatible with a wide angle of 170°, 3.7-mm instrument channel, and auxiliary water jet. A short transparent hood was fitted to the tip of the endoscope so that the whole ring could be seen in endoscopic view. Carbon dioxide was used for insufflation to decrease patient discomfort. We used a high-frequency electrosurgical generator (ESG-100; Olympus, Tokyo, Japan) with a peristaltic pump (AFU-100; Olympus, Tokyo, Japan). The energy setting used for incision and dissection was “forced coagulation 2” 30W, whereas “soft coagulation” 100W was used for haemostasis.

In our initial practice, we used the Flex Knife (KD-630L; Olympus, Tokyo, Japan) to perform the procedure. It had a loop-shaped tip that allowed easy control in any direction, as it was soft and flexible. Nonetheless, we found it difficult to adjust the length of the tip and there was frequent accumulation of debris around the tip. We then changed to Dual Knife (KD-650L; Olympus, Tokyo, Japan) with a fixed length (1.5 mm) and hence a more stable knife movement. More recently, we have used Flush Knife BT (DK2618JB/DK2618JN; Fujifilm, Saitama, Japan) for dissection. It has a ball tip of fixed length that touches a wider part of tissue and enhances haemostasis. The knife has a water jet channel and achieves two purposes: (1) it can wash away any tissue that accumulates around the tip, thereby maintaining the sharpness of the knife; and (2) submucosal normal saline injection can be performed without the need to change the instrument for further hyaluronate injection.

We used a mixture of 10% sodium hyaluronate (LG Chemical, South Korea) and 1:200 000 adrenaline saline at a ratio of 1:1.5. This solution was chosen for three reasons: the addition of adrenaline can produce a haemostasis effect; dilution of sodium hyaluronate made it less viscous and thus easier to inject; and it reduced the amount used of the relatively more expensive sodium hyaluronate.

All ESD procedures were performed by a single experienced endoscopist who had performed more than 500 therapeutic colonoscopic procedures and more than 200 laparoscopic colectomies. The ESD procedure was implemented by the endoscopist following completion of training on an animal model in the Second Master Workshop on Novel Endoscopic Technology & Endoscopic Submucosal Dissection in 2009 at Prince of Wales Hospital in Hong Kong, which was organised by the Department of Surgery, The Chinese University of Hong Kong (http://www.surgery.cuhk.edu.hk/events/2009-07-22-ESD.pdf). The workshop included both lecture sessions and hands-on sessions to perform ESD in a pig. The endoscopist had no experience in gastric or colorectal ESD. He received further overseas training in ESD in 2011 at Osaka Medical Center for Cancer and Cardiovascular Diseases in Japan as a clinical observer with hands-on animal model training.

With the patient initially lying in the left lateral position, a full colonoscopy was first performed to confirm and locate the site of pathology. Patients were then re-positioned such that the lesion was at an anti-gravitational position in the endoscopic view at 6 o’clock. This could be easily achieved by seeing the injected water pooling opposite to the lesion. In this position, the gravitational force aided in retracting the lesion away from the submucosal plane during dissection. We then injected 1:100 000 adrenaline saline at 1 cm distal to the lesion, aiming at the submucosal layer. This could be ascertained by seeing the formation of a dwell. With the injecting needle still in situ, the solution was then changed to the mixture of adrenaline saline and sodium hyaluronate to provide a precipitous elevation of sufficient height. After elevating the lesion, a mucosal incision was made proximal to the lesion. The mucosal incision was started at the proximal two thirds of the lesion. After mucosal incision, the submucosal plane was dissected with the submucosa dissected away from the muscle layer. Care was taken to manipulate the dissection plan parallel to the intestinal wall to prevent perforation. When a more than 1-mm diameter vessel was detected, it was coagulated using haemostasis forceps (Radial Jaw 4; Boston Scientific, US). When the flap was sufficient for retraction, the mucosal incision was completed. In case of perforation, the defect was closed with endoscopic clipping (EZ Clip; Olympus, Tokyo, Japan). The resected area was not closed as healing usually occurred in a few weeks without complications.16

All specimens were pinned on a piece of foam and fixed in formalin. Histological type, depth of invasion, as well as lateral and vertical resection margins were recorded. En-bloc resection was defined as one-piece resection of an entire lesion as observed endoscopically. R0 resection was defined as clear lateral and vertical resection margin.

All patients were allowed to resume a full diet on the same day. We performed no routine blood tests or imaging and patients were discharged the next day if there were no signs of perforation or haemorrhage. Postoperative haemorrhage was defined as clinical evidence of bleeding manifested by melena or haematochezia that required endoscopic haemostasis within 0 to 14 days of the procedure.11

All patients were followed up in clinic 2 weeks later to review the pathology report. Additional surgery would be offered in case of carcinoma with one of the following criteria: (1) margin involved; (2) >1 mm submucosal invasion; (3) positive lymphovascular permeation; (4) poorly differentiated adenocarcinoma, signet ring cell carcinoma, or mucinous carcinoma; or (5) high-grade tumour budding.17 Surveillance colonoscopy was performed 1 year after ESD.

All continuous variables were described as median and range. To study the learning curve, all patients were grouped chronologically into two periods: group 1 with cases 1 to 35; and group 2 with cases 36 to 71. Comparisons between non-parametric data were done with Mann-Whitney U test, while Chi squared test was used for categorical variables. A P value of <0.05 was considered statistically significant.

From March 2009 to December 2013, a total of 71 ESDs were performed. Characteristics of the patients are shown in Table 1.

There was one conversion to laparoscopic colectomy due to intra-operative bleeding that could not be controlled endoscopically (40 mm x 15 mm tumour at the transverse colon). For perforation, there were 11 (15.5%) perforations: eight in the colon and three in the rectum, all were noticed during the ESD. Endoscopic clipping was successful in 10 of the perforations. The median number of clips used was 2 (range, 1-6). One perforation required conversion to laparoscopic colectomy (30 mm x 5 mm tumour at the sigmoid colon). Both laparoscopic operations were uneventful and patients were discharged without any surgical complications. The overall morbidity rate was 16.9% including bleeding and perforation and there was no mortality. The median postoperative stay was 1 day (range, 0-11 days).

Of the 69 patients who completed the endoscopic procedure, en-bloc resection was successful in 56 (81.2%) patients, of whom 19 (27.5%) required additional snare (SD-210U-25, Snare Master; Olympus, Tokyo, Japan) to complete the en-bloc resection. Conversion to piecemeal resection by snare occurred in 14 (20.3%) patients.

In these 69 patients, the resection margin was unclear in 15 patients as it was too close to the cauterised edge. These, together with piecemeal resection, were classified as R1 (29 patients in total). R0 resection was achieved in 40 (58.0%) patients. The histopathological diagnosis was tubular adenoma for 26 (36.6%) tumours, tubulovillous adenoma for 28 (39.4%), villous adenoma for two (2.8%), serrated adenoma for six (8.5%), carcinoid for one (1.4%) with involved margin, intramucosal carcinoma for two (2.8%), and carcinoma with submucosal invasion for six (8.5%). The ESD procedure was considered curative for the two patients with intramucosal carcinoma. One patient who had submucosal carcinoma refused further treatment because of a subsequent diagnosis of primary lung cancer. All other patients with submucosal carcinoma had curative interval laparoscopic surgeries. There was no residual tumour and no lymph node involvement found in the surgical specimen for any of these patients. The patient with rectal carcinoid also underwent subsequent interval laparoscopic total mesorectal excision: the pathology was well-differentiated carcinoid with invasion to the muscularis propria. There was one lymph node metastasis out of 11 lymph nodes retrieved.

After excluding the six interval surgeries and two conversions to laparoscopic colectomy, the remaining 63 patients were offered colonoscopy surveillance of whom four refused and one defaulted from follow-up. Until August 2014, 51 patients had undergone surveillance colonoscopy and seven were awaiting colonoscopy. Recurrence of polyp occurred in seven (13.7%) out of 51 patients: three recurrences occurred after piecemeal resection, another three recurrences occurred after additional snare to complete the en-bloc resection. All three cases had uncertain margin due to proximity to the cauterised edge. In one patient, recurrence occurred after successful en-bloc resection by ESD, in which the deep margin was clear but the circumferential margin was not certain.

All patients were grouped chronologically into two periods: group 1 with cases 1 to 35; and group 2 with cases 36 to 71. The comparison between the two groups is shown in Table 2. The median procedure time per unit area of tumour improved significantly from 31.5 min/cm² to 21.5 min/cm² (P=0.032).

There were three (8.6%) perforations in group 1; one of them required conversion to laparoscopic colectomy. There were eight (22.2%) perforations in group 2; all managed successfully by endoscopic clipping. There was no significant difference in perforation rate (P=0.054). The intra-operative bleeding that required a conversion to laparoscopic colectomy also belonged to group 1.

For the 33 patients in group 1 who completed the endoscopic procedure, en-bloc resection was successful in 26 (78.8%), while 30 (83.3%) out of 36 patients in group 2 had successful en-bloc resection. This trend of improvement, however, did not reach statistical significance (P=0.15). Among these successful en-bloc resections, 15 (57.7%) out of 26 patients in group 1 and four (13.3%) out of 30 patients in group 2 required additional snare to complete the en-bloc resection (P<0.01).

R0 resection rate had improved significantly in group 2 despite a lower rate of snare application: 13 (39.4%) of 33 patients in group 1 had R0 resection, whereas in group 2, 27 (75.0%) of 36 patients had R0 resection (P<0.01).

The median postoperative stay was 1 day in both groups and was not significantly different (P=0.25). The median postoperative stay in patients with perforation in group 1 and group 2 was 1 day (range, 1-7 days) and 2 days (range, 1-11 days), respectively (P=0.73).

Colorectal ESD has been shown to be feasible and safe when performed at expert centres in Japan. In a recent prospective multicentre cohort study involving 1111 patients in Japan, the reported en-bloc resection rate was 88% and R0 resection rate was 89%.11 The total perforation rate was only 5.3% and postoperative bleeding was 1.5%.11 Nonetheless, these excellent results are largely reported from Japan. Colorectal ESD is technically difficult, as the lumen is narrow and angulated and the very thin wall presents a high risk of perforation. In Japan, endoscopists are first required to gain experience in gastric ESD, which is technically less demanding, before they move on to colorectal ESD.18 This is not possible in western and Asian countries outside Japan, where early gastric cancer is much less prevalent. In a recent review of 82 rectosigmoid ESD from a tertiary centre in Germany, the en-bloc resection rate and R0 resection rate was only 81.6% and 69.7%, respectively.19 This reflects the difficulty in generalising this novel procedure in other counties outside Japan, and an even greater challenge to low-volume district centres that lack local expertise.

The low case volume and the absence of expertise in western countries leads to the development of untutored colorectal ESD when it is impossible to have a step-up approach in ESD training starting from the stomach before proceeding to colon. The reported learning curve for untutored colorectal ESD has an acceptable outcome, however. Berr et al20 reported a case series of 48 colorectal ESDs with 76% en-bloc resection, 14% perforation, and 4% requirement for surgical intervention. This compared favourably with results in Japan.11 Another learning curve study of colorectal ESD found procedure time could be significantly shorter after 25 procedures.21

With a similar situation in Hong Kong, this study showed that untutored colorectal ESD is safe and feasible. Results demonstrated obvious improvement after 35 procedures, as evidenced by the significant reduction in combined ESD and EMR with snare from 45.5% to 11.1%. Procedure time per unit area of tumour as well as R0 resection rate also significantly improved after 35 cases. Although there were more perforations in group 2, it did not determine adverse outcome. None of the perforations in group 2 required conversion to laparoscopic colectomy. There was no significant impact on hospital stay. The higher perforation rate in group 2 may reflect a second learning curve to perform a complete ESD procedure without the assistance of a hybrid technique to achieve a reasonable R0 resection rate. A perforation rate comparable with the Japanese series11 is expected in the third tranche of 35 patients.

Contrary to perforations in traditional therapeutic colonoscopies, all perforations in ESD were only 1 to 2 mm in size and were noticed intra-operatively, thus immediate endoscopic repair was possible. No patient required surgical intervention solely for treatment of perforation. The only conversion to laparoscopic colectomy in the initial learning curve aimed to offer one-stop treatment rather than treating perforations. In a multicentre review of colonoscopic perforations by Teoh et al,22 43 (0.113%) perforations were found in 37 971 colonoscopies. Only seven (43.8%) out of 16 therapeutic colonoscopic perforations were noticed during the endoscopic procedure. The mean size of perforation was 0.98 cm. The overall 30-day morbidity and mortality rate was 48.7% and 25.6%, respectively and the stoma rate was 38.5%. This showed clearly that surgical outcome was much worse in conventional colonoscopic perforations compared with perforations in ESD, despite a much higher perforation rate of 15.7% in ESD group.22

After implementation of the ESD service, the following were noted:
(1) Venue of procedure—operating theatre was chosen instead of an endoscopic unit to enable conversion to conventional laparoscopic colectomy without the need to change location as well as the ready availability of an anaesthetist to give conscious sedation.
(2) Mode of anaesthesia—in the first few patients, we performed ESD under general anaesthesia for patient comfort and in the event conversion to laparoscopic colectomy was necessary. Positioning of patients was clumsy particularly when a prone position was needed to perform the procedure. Subsequently conscious sedation by an anaesthetist was used instead. Patients could follow instructions for positioning and deeper sedation could be achieved if necessary. There were no complaints from patients about any discomfort during the procedure.
(3) Choice of injecting agents—albumin 20% (Albumex 20; CSL, Australia) was used as submucosal injecting agent in the first few patients when sodium hyaluronate was not available. Albumin 20% has both a good cushioning effect without any inflammatory effect and the cost was much cheaper at HK$2.7/mL compared with commercially available sodium hyaluronate at HK$68/mL.23 Yet sodium hyaluronate has the longest lasting cushioning effect among all injecting agents. We recommend its use whenever available.
(4) A mixture of adrenaline saline and sodium hyaluronate was favourable for the assistant to inject and a lesser volume of sodium hyaluronate was required. Moreover, there were no instances of postoperative haemorrhage, although it was difficult to conclude whether this was due to the addition of adrenaline saline.
(5) Endoscopic technique—in our initial practice, we performed a full circumferential mucosal incision before submucosal dissection. We noticed it was technically more difficult compared with a two-third circumferential incision, because firstly, the submucosal elevation was lost quickly due to faster leakage of injecting agent, and secondly, it was difficult to retract the lesion at the end of dissection, and we had to complete the en-bloc resection with snare. After changing to two-third circumferential incision in the second period of the study, the need for additional snare to complete the en-bloc resection was significantly decreased (from 57.7% to 13.3%).
(6) Postoperative management—it was feasible and safe to resume diet immediately after the procedure and discharge patients the day following ESD without the need for routine blood taking or imaging. In one study from Japan, abdominal computed tomography was performed on day 1 and blood tests were carried out for 2 consecutive days. Oral intake was gradually stepped up and patients were discharged 5 days after ESD.10 In contrast, we allowed full diet on the same day after ESD and did not perform computed tomography or blood tests routinely. Our overall median postoperative stay was 1 day. Further development of ESD as a day procedure can be explored.

There are a number of limitations in this study. First, this was the learning curve of a single endoscopist and 35 procedures may not be a typical number required for such a learning curve. Second, the inclusion criteria for colorectal ESD were less strict than those in Japan. Lesion less than 2 cm that could not be removed en bloc would be subjected to ESD instead of piecemeal resection. These kinds of lesion were expected to be easier with shorter operating time. Third, this was a retrospective comparison of two chronological groups which might not have been directly comparable. Lastly, the 7.9% patient default rate may lead to underestimation of recurrence in this series.

Untutored colorectal ESD at a low-volume centre was an option in the absence of enough experts to supervise the procedure. Training on an animal model and clinical observation of real-time demonstrations was useful to start ESD without supervision. A cut-off at 35 procedures showed an acceptable R0 resection rate at a significantly improved procedure time per unit area. There was a second learning curve to achieve a complete ESD procedure without EMR at a higher perforation rate. Colorectal ESD performed by a colorectal surgeon enables any complications to be managed by the same operator, or any lesion unresectable by ESD to be surgically removed. It was not necessary to first perform gastric ESD as the start of ESD training. When more endoscopists have gained experience in colorectal ESD, a structured training programme with accreditation can be established.

All authors have disclosed no conflicts of interest.

1. Tanaka S, Haruma K, Oka S, et al. Clinicopathological features and endoscopic treatment of superficially spreading colorectal neoplasms larger than 20 mm. Gastrointest Endosc 2001;54:62-6. Crossref

2. Walsh RM, Ackroyd FW, Shellito PC. Endoscopic resection of large sessile colorectal polyps. Gastrointest Endosc 1992;38:303-9. Crossref

3. Uraoka T, Fujii T, Saito Y, et al. Effectiveness of glycerol as a submucosal injection for EMR. Gastrointest Endosc 2005;61:736-40. Crossref

4. Yamamoto H, Kawata H, Sunada K, et al. Successful en-bloc resection of large superficial tumors in the stomach and colon using sodium hyaluronate and small-caliber-tip transparent hood. Endoscopy 2003;35:690-4. Crossref

5. Fujishiro M, Yahagi N, Kakushima N, et al. Outcomes of endoscopic submucosal dissection for colorectal epithelial neoplasms in 200 consecutive cases. Clin Gastroenterol Hepatol 2007;5:678-83. Crossref

6. Saito Y, Fukuzawa M, Matsuda T, et al. Clinical outcome of endoscopic submucosal dissection versus endoscopic mucosal resection of large colorectal tumors as determined by curative resection. Surg Endosc 2010;24:343-52. Crossref

7. The Paris endoscopic classification of superficial neoplastic lesions: esophagus, stomach, and colon: November 30 to December 1, 2002. Gastrointest Endosc 2003;58(6 Suppl):S3-43.

8. Watanabe T, Itabashi M, Shimada Y, et al. Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines 2010 for the treatment of colorectal cancer. Int J Clin Oncol 2012;17:1-29. Crossref

9. Yoda Y, Ikematsu H, Matsuda T, et al. A large-scale multicenter study of long-term outcomes after endoscopic resection for submucosal invasive colorectal cancer. Endoscopy 2013;45:718-24. Crossref

10. Yoshida N, Wakabayashi N, Kanemasa K, et al. Endoscopic submucosal dissection for colorectal tumors: technical difficulties and rate of perforation. Endoscopy 2009;41:758-61. Crossref

11. Saito Y, Uraoka T, Yamaguchi Y, et al. A prospective, multicenter study of 1111 colorectal endoscopic submucosal dissections (with video). Gastrointest Endosc 2010;72:1217-25. Crossref

12. Taku K, Sano Y, Fu KI, et al. Iatrogenic perforation associated with therapeutic colonoscopy: a multicenter study in Japan. J Gastroenterol Hepatol 2007;22:1409-14. Crossref

13. Sano Y, Muto M, Tajiri H, et al. Optical/digital chromoendoscopy during colonoscopy using narrow-band image system. Dig Endosc 2005;17(Suppl):S43-8. Crossref

14. Ikematsu H, Matsuda T, Emura F, et al. Efficacy of capillary pattern type IIIA/IIIB by magnifying narrow band imaging for estimating depth of invasion of early colorectal neoplasms. BMC Gastroenterol 2010;10:33. Crossref

15. Machida H, Sano Y, Hamamoto Y, et al. Narrow-band imaging in the diagnosis of colorectal mucosal lesions: a pilot study. Endoscopy 2004;36:1094-8. Crossref

16. Iguchi M, Yahagi N, Fujishiro M, et al. The healing process of large artificial ulcers in the colorectum after endoscopic mucosal resection [abstract]. Gastrointest Endosc 2003;57:AB226.

17. Asano M. Endoscopic submucosal dissection and surgical treatment for gastrointestinal cancer. World J Gastrointest Endosc 2012;4:438-47. Crossref

18. Tanaka S, Tamegai Y, Tsuda S, Saito Y, Yahagi N, Yamano HO. Multicenter questionnaire survey on the current situation of colorectal endoscopic submucosal dissection in Japan. Dig Endosc 2010;22 Suppl 1:S2-8. Crossref

19. Probst A, Golger D, Anthuber M, Märkl B, Messmann H. Endoscopic submucosal dissection in large sessile lesions of the rectosigmoid: learning curve in a European center. Endoscopy 2012;44:660-7. Crossref

20. Berr F, Wagner A, Kiesslich T, Friesenbichler P, Neureiter D. Untutored learning curve to establish endoscopic submucosal dissection on competence level. Digestion 2014;89:184-93. Crossref

21. Białek A, Pertkiewicz J, Karpińska K, Marlicz W, Bielicki D, Starzyńska T. Treatment of large colorectal neoplasms by endoscopic submucosal dissection: a European single-center study. Eur J Gastroenterol Hepatol 2014;26:607-15. Crossref

22. Teoh AY, Poon CM, Lee JF, et al. Outcomes and predictors of mortality and stoma formation in surgical management of colonoscopic perforations: a multicenter review. Arch Surg 2009;144:9-13. Crossref

23. ASGE Technology Committee, Kantsevoy SV, Adler DG, Conway JD, et al. Endoscopic mucosal resection and endoscopic submucosal dissection. Gastrointest Endosc 2008;68:11-8. Crossref

Osteoporosis is a skeletal disorder characterised by low bone mass and micro-architectural deterioration of bone tissue that results in compromised bone strength and a predisposition to fracture.1 Childhood and adolescence are critical periods for bone mineralisation. Peak bone mineral density (BMD) achieved by early adulthood determines the risk of pathological fracture. Bone mass can be objectively measured by BMD and is correlated with risk of osteoporotic fracture.2

Dual-energy X-ray absorptiometry (DXA) scanning enables cost-effective quantitative measurement of BMD. Such methodology is the gold standard because it can detect bone mineral loss of 2% to 5%. Abnormal homeostasis of vitamin D, calcium, and phosphorous may lead to imbalanced osteoclastic and osteoblastic dynamics that will result in osteopenia and even osteoporosis. There have been reports from large- and small-scale studies of problems in bone mineralisation and vitamin D or calcium metabolism in patients with cerebral palsy (CP).3 4 Dietary restrictions, oromotor dysfunction, malabsorption, and limited sunlight exposure may lead to poor nutrition, low calcium intake, or altered vitamin D metabolism. These in turn contribute to poor mineralisation. Limited weight-bearing during the period of skeletal growth may also lead to reduced BMD. Chronic therapy with antiepileptic drugs (AEDs) may cause hypocalcaemia, hypophosphataemia, raised serum alkaline phosphatase, elevated serum parathyroid hormone, reduced biologically active vitamin D metabolites, rickets, and osteomalacia. The mechanism is unclear, however.5

A study revealed that AED use was associated with reduced BMD in 35 pairs of twin adults, and the effect was more marked in those prescribed enzyme-inducing AEDs, for example, phenytoin and phenobarbitone.6 A decrease in BMD (in g/cm²) of 6.4% and 4.6% at the lumbar spine and femoral neck respectively in their twin control pair was detected. For valproate, the effect was not consistently demonstrated.7 8

Chronic treatment with AEDs was observed to be significantly correlated with a lower BMD in 96 ambulatory children and young adults with epilepsy with a mean Z-score of -1.23 compared with 0.16 of a control group.9 In our retrospective study of 109 CP children in 2006, however, we could not demonstrate any association of AEDs with increased fracture rate in non-ambulatory CP children.10 In children with CP and a Gross Motor Function Classification System (GMFCS) level of 4 to 5 (bed-bound), the possible impact of anticonvulsants on BMD was not clearly demonstrated.

The primary objective of this study was to evaluate the effect of AEDs on BMD in non-ambulatory children with CP. The secondary objective was to identify the risk factors for low BMD in this group of children.

Patients at the Developmental Disabilities Unit (DDU) of Caritas Medical Centre, Hong Kong were enrolled in the study from 1 October 2012 to 30 September 2013. The DDU is the largest long-term care facility for children with severe developmental disabilities and special health care needs in Hong Kong. Among the 100 residents, over 90% were non-ambulatory and over 50% had CP.

The inclusion criteria in this study were: (1) children and adolescents aged between 5 and 19 years; (2) GMFCS level 4 or 5; (3) the presence of spastic or mixed spastic dyskinetic CP; and (4) prescribed AED for more than 2 years.

The control group without anticonvulsant therapy comprised concomitant non-ambulatory DDU residents with CP who were not prescribed AEDs for more than 2 years prior to recruitment to the study.

The exclusion criteria included (1) GMFCS level of 1 to 3; (2) patients with underlying hepatic or renal disease; (3) presence of disease primarily involving bone metabolism or a family history of bone metabolic disorders; (4) known thyroid or parathyroid disease; (5) history of pathological long bone fracture; (6) history of chronic diarrhoea or malabsorption; (7) long-term intake of the following medications: non-physiological dose of glucocorticoid, anabolic steroid, vitamin A, non-steroidal anti-inflammatory drugs, bisphosphonates, thiazide diuretics, or calcitonin.

Coppola et al9 reported a mean BMD Z-score of -1.23 and 0.16 in 96 ambulatory epileptic patients and 63 controls, respectively. In 2012, the same group showed the mean Z-score to be -1.69 (n=47) and -0.83 (n=40) respectively in mentally retarded children with CP, with and without epilepsy.11 Based on these findings, we proposed a difference in Z-score of ≥1 to indicate a clinically significant reduction in BMD. In order to detect a Z-score difference of 1 with power of 0.8, 16 subjects were recruited in each group.12

Upon entry, data were collected for age, gender, body weight, body height, body mass index (BMI), presence of contractures, skinfold thickness (triceps, lower biceps, and subscapular), and pubertal stage according to Tanner’s classification. Weight-for-age Z-score was calculated based on referenced data from a 1993 territory-wide growth survey of Hong Kong children.13 Blood taking was performed on the same day of DXA scan to evaluate calcium, phosphorous, alkaline phosphatase, transaminase, total protein, urea, creatinine, serum osmolality, and total cell count. A spot urine calcium-to-creatinine ratio (mmol/mmol) and urine calcium-to-osmolality ratio (mg/L:mosmo/kg) was also determined on the same day. Current dietary information was reviewed and daily calcium and vitamin D content were calculated by a dietitian.

In this study, BMD was determined by DXA of total body less head (TBLH) using a Lunar Prodigy Advance DXA bone densitometer (GE Healthcare, Madison [WI], US) with the latest Chinese children total BMD database installed.14 15 All subjects had BMD measured by the same validated densitometer to eliminate measurement error. Individual BMD value was expressed as g/cm² and Z-score. Z-score is the number of standard deviations of the patient’s BMD above or below the average age- and sex-matched reference value. For the paediatric age-group, the current definition for osteoporosis includes BMD Z-score of ≤–2.0 adjusted for age, gender, and body size plus a clinically significant fracture defined in the International Society for Clinical Densitometry (ISCD) 2007 official positions.16 The head is disproportionately large in young children and may mask deficits at other skeletal sites, thus TBLH (subtotal/TBLH) BMD, which was recommended by ISCD 2007 official positions for the evaluation of child’s bone health, was used in this study. For children with CP, TBLH BMD is more feasible and provides a holistic picture for BMD estimation.

In bivariate analysis, normally and non-normally distributed data were analysed by independent sample t tests and Mann-Whitney U tests, respectively. Categorical data were compared using Chi squared and Fisher’s exact tests. Clinically relevant covariates (presence of puberty, daily calcium intake, and number of AED use) and the covariate (weight-for-age Z-score) approaching statistical significance (P<0.05) in bivariate analysis were entered into multivariable analysis (Enter Method). Statistical significance was defined as two-tailed probability below 0.05.

Data of TBLH were split into low-BMD group versus normal-BMD group—those with TBLH Z-score of ≤–2.0 were considered to have low BMD. Baseline characteristics of the low- and normal-BMD groups were compared. Multivariable logistic regression analysis was used to examine the independent effect of puberty, weight-for-age Z-score (gender adjusted), number of AEDs, and daily calcium intake on TBLH Z-score.

All statistical analyses were performed using the Statistical Package for the Social Sciences (Windows version 13.0; SPSS Inc, Chicago [IL], US).

The study was approved by the Ethics Committee of Kowloon West Cluster of Hospital Authority, Hong Kong. Informed consent was obtained from respective parents or the legal guardians from the Social Welfare Department, Government of the Hong Kong Special Administrative Region.

The subjects were enrolled from 1 October 2012 to 30 September 2013 in DDU. Sixty-two children with CP aged 5 to 19 years were identified. Informed consent was obtained from a legal guardian for 44 patients of whom 12 were excluded from study—four had a history of fracture, two had severe deformity making DXA technically difficult, one had poorly controlled asthma and required long-term steroid therapy, and five had deranged liver or renal function. Of the remaining 32 children and adolescents, 18 (6 males, 12 females) aged 5.0 to 19.5 (mean ± standard deviation, 13.8 ± 4.7) years comprised the study group. The control group consisted of 14 children and adolescents (6 males, 8 females) aged 7.0 to 19.1 (mean, 16.4 ± 3.0) years.

Among the 18 subjects in the study group, eight were prescribed one AED and 10 were prescribed two or more. Sodium valproate was prescribed to eight children, carbamazepine to three, topiramate to three, phenobarbitone to five, phenytoin to one, and lamotrigine to four. In view of the limited population size, the impact of individual anticonvulsants was not analysed.

The two groups showed no significant difference in age, gender, BMI, weight-for-age Z-score, CP type, daily calcium intake, or daily vitamin D intake (Table 1).

Twelve (67%) patients in the study group were in puberty, compared with 13 (93%) patients in the control group (P=0.104). Sixteen (89%) patients in the study group were tube feeders compared with four (29%) in the control group (P=0.001). Among the oral feeders, all were reported by their caretaker to have a satisfactory dietary intake. Mean weight-for-age Z-score showed no significant difference between the study and control groups (P=0.226) [Table 1]. There was also no significant difference in the TBLH Z-scores (P=0.989; Table 1) or biochemical parameters of bone metabolism between the two groups (Table 2).

Low BMD was defined as TBLH Z-score of ≤–2.0 and was evident in 18 (56%) patients. When comparing groups with low and normal BMD, there was no significant difference on univariate analysis in gender, age, CP, BMI, feeding mode, use of AEDs, puberty, or daily calcium and vitamin D intake (Table 3). Weight-for-age Z-score was significantly lower (-2.57 vs -1.12; P=0.001) in the low BMD group (Table 3).

When covariates including weight-for-age Z-score, presence of puberty, daily calcium intake, and number of AEDs used were analysed by multiple logistic regression, higher weight-for-age Z-score (adjusted odds ratio [OR]=0.015; 95% confidence interval [CI], 0.001-0.390) and presence of puberty (adjusted OR=0.027; 95% CI, 0.001-0.948) remained significant independent predictors for occurrence of a relatively normal BMD (ie TBLH Z-score >2.0). Hosmer-Lemeshow goodness of fit test (P=0.315) was non-significant, indicating a model prediction that was not significantly different to observed values. Nagelkerke R² was 0.677, again signifying a relatively well-fitting model (Table 4).

In the present study, DXA evaluation in non-ambulatory children with CP, with or without AED use, revealed a low BMD in 56% of those who underwent DXA. This is concordant with the findings in previous studies4 17 18 19 with prevalence of low BMD ranging from 27% to 77%. Low BMD is the proximate cause of low-energy fracture in this group of children. Other factors such as stiff joints, poor balance leading to falls, and violent seizures may also contribute. Nearly 20% of such children have sustained a femoral fracture at some point.20

The impact of AED on BMD was unclear in non-ambulatory children with CP. The physician often faces a dilemma in optimising anticonvulsant treatment in this group of children who are already at high risk of developing osteoporotic long bone fracture.

Early studies revealed that treatment with enzyme-inducing anticonvulsants—for example, phenytoin and phenobarbitone—may induce catabolism of 25-hydroxyvitamin D, leading to rickets.21 22 Recent research on the effect of AED on BMD is inconclusive, and the results are often confounded by methodological flaws. In a cross-sectional study, Farhat et al5 measured the BMD in 71 ambulatory subjects who were prescribed AED for more than 6 months. Reduced BMD was found in adults (n=42) but not children (n=29). There was no control group and the findings were only compared with paediatric data from the manufacturer, rendering it difficult to detect any small but significant difference secondary to AED treatment. Ecevit et al23 measured femoral neck BMD in 31 healthy children and 33 subjects with idiopathic epilepsy treated with either carbamazepine (n=17) or valproate (n=16). Valproate but not carbamazepine monotherapy was associated with a significant reduction in BMD. In this study, however, more children had attained puberty in the control group (n=13, 93%) than in the AED group (n=12, 67%), which may have confounded the results. Coppola et al9 compared the BMD of 96 children with epilepsy alone or in association with CP and/or mental retardation against a control group of 63 healthy ambulatory subjects. While univariate analysis illustrated a reduced BMD Z-score in the AED group, no difference was found after adjustment for ambulatory status, mental retardation, lack of physical activity, and BMI in multivariate analysis.

In this study, we focused on non-ambulatory children with CP and severe mental retardation. This is the population most vulnerable to pathological long bone fracture, with a lifetime risk estimated to be 20%. This group of children is also prone to develop refractory epilepsy that requires multiple AEDs at high doses. By recruiting concomitant residents from the same facility as the control group, we minimised the confounding effect of factors that may contribute to low BMD, namely ambulatory status, physical activity, and nutritional status. Our findings support that long-term use of AED in this group of children does not have any significant adverse effect on BMD. Optimal pharmacological treatment can be pursued without jeopardising bone mineralisation.

In multivariable analysis, there was a significant association of BMD Z-score with the presence of puberty. This may be due to the physiology and time frame for bone mineralisation that surges in puberty and reaches a peak at 20 years of age. Variation in time of starting puberty was a confounding factor as BMD Z-score was adjusted for gender and age only.

This study was carried out in a single institution for disabled children in Hong Kong. Because of the need for sedation and exposure to radiation, guardians of children with more severe neurological disease tended not to consent for the study. Moreover, DXA could not be properly performed in patients with severe deformity. These children were not included in the study.

An inadequate number of patients in each group may have impacted the power of this study. Nevertheless, the result showed a strong association for weight-for-age Z-score with TBLH BMD. We believe the association was highly statistically significant.

Children with CP are at risk of malnutrition that has a significant impact on linear growth, wound recovery, motor function, and even survival. In addition to poor linear growth, malnourished children with severe CP are likely to have poor bone mineralisation leading to painful pathological fracture. In a retrospective study of 107 CP children (90 non-ambulatory), weight-for-age Z-score proved to be the best predictor of BMD Z-score.24 In the present study, we demonstrated a significant correlation of low weight-for-age Z-score with low BMD, with a relatively well-fitting model after adjusting for pubertal stage, calcium intake, and AED use. This is concordant with our previous findings in the same population in which there was a significant protective effect of increase in weight-for-age Z-score in reducing fracture risk (adjusted OR=0.41).10 A recent review of 32 cross-sectional, cohort, case-control, and randomised controlled trials suggested that reduced bone density, impaired bone growth, and vitamin D deficiency may be seen in children treated with anticonvulsants.25 In our study, individual vitamin D daily intake was calculated by dietitians. Sun exposure was limited in the institutional setting. Thus, we assumed that the calculated daily oral vitamin D intake correlated well with the serum vitamin D level. Although the daily vitamin D intake was similar in the low- and normal-BMD groups (P=0.297) as well as the study and control groups (P=0.702), 10 out of 32 subjects in the study consumed less than the recommended daily vitamin D intake (400 IU/day). To improve BMD and decrease fracture risk, regular nutritional assessment is necessary to identify malnourished children. Supplementation of vitamin D and calcium should be considered if the calculated daily need cannot be met. A nutritional rehabilitation programme should be implemented to optimise weight-for-age Z-score. Early gastrostomy tube placement should be considered in children with poor oral feeding. In children with weight-for-age Z score of ≤–2.0, serial BMD measurements can help to identify severe osteoporosis before fracture occurs. This window of time allows implementation of intensive nutritional rehabilitation, pharmacological intervention, and precautions in handling to prevent fracture occurrence.

Use of AED in non-ambulatory children with CP is unlikely to pose a detrimental impact on BMD. A low weight-for-age Z-score is a significant independent risk factor predictive of low BMD. Optimising nutritional status in this group of children is paramount to improving bone mineralisation and thus decreasing pathological fracture.

The authors would like to thank Ms Geraldine Ng, dietitian of Caritas Medical Centre, for her arduous effort to assess dietary calcium and vitamin D intake for our study patients.

All authors have disclosed no conflicts of interest.

1. Houlihan CM, Stevenson RD. Bone density in cerebral palsy. Phys Med Rehabil Clin N Am 2009;20:493-508. Crossref

2. Marshall D, Johnell O, Wedel H. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ 1996;312:1254-9. Crossref

3. Lee JJ, Lyne ED, Kleerekoper M, Logan MS, Belfi RA. Disorders of bone metabolism in severely handicapped children and young adults. Clin Orthop Relat Res 1989;245:297-302. Crossref

4. Henderson RC, Lark RK, Gurka MJ, et al. Bone density and metabolism in children and adolescents with moderate to severe cerebral palsy. Pediatrics 2002;110:e5. Crossref

5. Farhat G, Yamout B, Mikati MA, et al. Effect of antiepileptic drugs on bone density in ambulatory patients. Neurology 2002;58:1348-53. Crossref

6. Petty SJ, Paton LM, O’Brien TJ, et al. Effect of antiepileptic medication on bone mineral measures. Neurology 2005;65:1358-65. Crossref

7. Sheth RD, Wesolowski CA, Jacob JC, et al. Effect of carbamazepine and valproate on bone mineral density. J Pediatr 1995;127:256-62. Crossref

8. Triantafyllou N, Lambrinoudaki I, Armeni E, et al. Effect of long-term valproate monotherapy on bone mineral density in adults with epilepsy. J Neurol Sci 2010;290:131-4. Crossref

9. Coppola G. Fortunato D, Auricchio G, et al. Bone mineral density in children, adolescents, and young adults with epilepsy. Epilepsia 2009;50:2140-6. Crossref

10. Ko CH, Tse PW, Chan AK. Risk factors of long bone fracture in non-ambulatory cerebral palsy children. Hong Kong Med J 2006;12:426-31.

11. Coppola G, Fortunato D, Mainolfi C, et al. Bone mineral density in a population of children and adolescents with cerebral palsy and mental retardation with or without epilepsy. Epilepsia 2012;53:2172-7. Crossref

12. Portney LG, Watkins MP. Foundations of clinical research applications to practice. 3rd ed. Upper Saddle River, NJ: Pearson/Prentice Hall; 2009: 832-47.

13. Leung SS, Lau JT, Tse LY, Oppenheimer SJ. Weight-for-age and weight-for-height references for Hong Kong children from birth to 18 years. J Paediatr Child Health 1996;32:103-9. Crossref

14. Xu H, Chen JX, Gong J, et al. Normal reference for bone density in healthy Chinese children. J Clin Densitom 2008;10:266-75. Crossref

15. Guo B, Xu Y, Gong J, Tang Y, Xu H. Age trends of bone mineral density and percentile curves in healthy Chinese children and adolescents. J Bone Miner Metab 2013;31:304-14. Crossref

16. Lewiecki EM, Gordon CM, Baim S, et al. International Society for Clinical Densitometry 2007 Adult and Pediatric Official Positions. Bone 2008;43:1115-21. Crossref

17. Hartman C, Brik R, Tamir A, Merrick J, Shamir R. Bone quantitative ultrasound and nutritional status in severely handicapped institutionalized children and adolescents. Clin Nutr 2004;23:89-98. Crossref

18. King W, Levin R, Schmidt R, Oestreich A, Heubi JE. Prevalence of reduced bone mass in children and adults with spastic quadriplegia. Dev Med Child Neurol 2003;45:12-6. Crossref

19. Ali O, Shim M, Fowler E, Cohen P, Oppenheim W. Spinal bone mineral density, IGF-1 and IGFBP-3 in children with cerebral palsy. Horm Res 2007;68:316-20. Crossref

20. Sturm PF, Alman BA, Christie BL. Femur fractures in institutionalized patients after hip spica immobilization. J Pediatr Orthop 1993;13:246-8.

21. Crosley CJ, Chee C, Berman PH. Rickets associated with long-term anticonvulsants therapy in a pediatric outpatient population. Pediatrics 1975;56:52-7.

22. Keck E, Gollnick B, Reinhardt D, Karch D, Peerenboom H, Krüskemper HL. Calcium metabolism and vitamin D metabolite levels in children receiving anticonvulsant drugs. Eur J Pediatr 1982;139:52-5. Crossref

23. Ecevit C, Aydoğan A, Kavakli T, Altinöz S. Effect of carbamazepine and valproate on bone mineral density. Pediatr Neurol 2004;31:279-82. Crossref

24. Henderson RC, Kairalla J, Abbas A, Stevenson RD. Predicting low bone density in children and young adults with quadriplegic cerebral palsy. Dev Med Child Neurol 2004;46:416-9. Crossref

25. Vestergaard P. Effects of antiepileptic drugs on bone health and growth potential in children with epilepsy. Pediatr Drugs 2015;17:141-50. Crossref

According to the Family Planning Knowledge, Attitude and Practice in Hong Kong Survey 2012 among Hong Kong couples,1 the male condom was the most popular contraceptive. The proportion of couples who used a male condom doubled from 32.2% in 1987 to 69.6% in 2012. Combined oral contraceptive (COC) pill was the second most common form of contraception, though the proportion of women using a COC pill declined from 20.3% in 1987 to 10.8% in 2012. The failure rate of the male condom when used correctly is 6 times higher than that for the COC pill.2 Although the low-dose COC pill has a low incidence of complications, high efficacy, and many non-contraceptive benefits, relatively few women use it in Hong Kong. The report of the United Nations world contraceptive patterns 2013 estimated that the prevalence of pill use in Hong Kong women aged 15 to 49 years was 6.7%, which is much lower than other countries with similar development, wealth, and culture such as Australia (30.0%), Canada (21.0%), Singapore (10.0%), the UK (28.0%), and the US (16.3%).3 Unlike these countries, the COC pill is not a prescription drug in Hong Kong. Women can buy a low-dose COC pill that contains either 30-µg or 20-µg ethinylestradiol and one of the progestogens: levonorgestrel, gestodene, desogestrel, or drospirenone at any of the large-chain personal health and beauty retailers or pharmacy stores. All pills have similar efficacy. Their failure rate is 0.3% within the first year of perfect use.2 Low-dose pills are safer, better tolerated, and have equal or higher efficacy than high-dose pills that contain 50-µg ethinylestradiol.

With 70% of couples in Hong Kong using the male condom,1 the demand for abortion due to failed contraception cannot be ignored. It was shown that 77.4% of women who underwent an abortion were using contraception during the index pregnancy and 51.2% of them were using a male condom.1 The number of legal abortions in Hong Kong has reduced from 25 363 in 1995 to 10 359 in 2014 (personal communication, Department of Health), though the number of abortions carried out across the border is unknown. According to the results from the serial 5-yearly territory-wide family planning survey,1 the proportion of married women who went to China for their last abortion increased from 24.3% in 1992 to 47.2% in 2012. Given the limited resources assigned to abortion in public hospitals, women have to resort to the more expensive legal abortion service in private hospitals or the Family Planning Association of Hong Kong (FPAHK). The FPAHK performs 3000 medical and surgical first-trimester abortions each year and has reached its full service capacity. There is a need to further reduce unplanned pregnancies and abortion in Hong Kong. One plausible solution is to encourage more women to use more effective contraception such as the combined hormonal contraceptive pill, progestogen-only contraceptives, intrauterine contraceptive device, or sterilisation. The failure rate of these effective contraceptives when used correctly is <1% in the first year of use.2

Studies have shown that identifying women’s perspective can help doctors understand their motive to choose one contraceptive over another.4 One study identified personal choices, local factors, women’s perceived safety, effectiveness, and convenience of the method as determinants of contraceptive choice.5 Among the effective contraceptives available in Hong Kong, the COC pill is the most accepted. We performed this survey to determine the motivators and barriers to COC pill use.

The Family Planning Association of Hong Kong invited ESDlife to host the survey that was open to its female members aged between 18 and 45 years. The questionnaire was designed by the investigators. ESDlife is an online lifestyle media in Hong Kong. It is a joint venture between the Hong Kong SAR Government and a commercial firm that began in 2000 with the aim of providing e-government and e-commerce services. With the establishment of the Government’s own website in 2008, all government services have migrated to the official website and ESDlife remains a solely commercial portal. As of 2 January 2015, it had 297 152 members of whom 64.4% were female. Among the female members, 89.0% were within our target age range: 32.3% were 30-34 years old, 23.3% were 35-39 years old, 19.3% were 25-29 years old, 10.3% were 40-45 years old, and 3.8% were 18-24 years old.

ESDlife sent out 100 000 invitations randomly to its female members aged between 18 and 45 years on 21 April 2015 and invited them to participate in this online survey between 21 April 2015 and 20 May 2015. There were 16 questions that explored the basic demographic characteristics of respondents (6), their contraceptive choice (3), and their motivators and barriers to COC use (7). Invited members entered the survey via a link and those who had not used any contraception in the previous 12 months were screened out by the first question. Only eligible subjects could proceed with the survey. They could stop at any question and the questionnaire would be voided. To encourage participation, a $50 supermarket coupon was given to every 10th respondent via ESDlife. This study was reviewed and approved by the Health Services Subcommittee and Ethics Panel of the FPAHK.

Data analyses were accomplished using the Statistical Package for the Social Sciences (Windows version 22.0; SPSS Inc, Chicago [IL], US). Descriptive statistics were presented. Bivariate Chi squared test was performed to analyse the demographic characteristics that predicted COC pill use.

During the survey period, only completed questionnaires were captured by the system so the number of incomplete questionnaires was unknown. A total of 1566 women completed the survey within the 1-month period, 271 were screened out by Question 1 because they had not used regular contraception in the past 12 months and 1295 questionnaires were analysed. The response rate was 1.57%. The median age of the respondents was 32 years (interquartile range, 29-36 years). Half of them (50.7%) had a university education, 20.5% had a post-secondary education (diploma or associate degree), and 28.3% had a secondary education. The majority (65.5%) were married, 29.0% were unmarried, 3.3% were cohabiting, 2.1% had separated or divorced, and 0.2% were widowed. Over half of the respondents were nulliparous (56.7%) and had no plan for pregnancy (52.3%). They usually purchased contraceptives from a chain of personal health and beauty retailers (52.8%), convenience store and supermarket (43.9%), or pharmacy (23.6%). They usually sought contraceptive information from an online health website (46.5%), online forum (40.1%), gynaecologists (27.8%), or family planning clinic (21.6%). A summary of the socio-demographic characteristics is shown in Table 1.

Among the 1295 respondents, 453 (35.0%) had used more than one type of contraceptive in the previous 12 months. The contraceptive choices of the whole group were: male condom (76.1%), coitus interruptus (20.9%), safe period (16.2%), and COC pill (12.6%) [Table 1]. The contraceptive choices of the 986 male condom users were further analysed to estimate their risk of unplanned pregnancy. Among them, 598 (60.6%) used a male condom alone, 295 (29.9%) also used other less effective contraceptives such as a female condom, safe period, and coitus interruptus but whether they used them all together during coitus or switched between these contraceptives was unknown. Therefore, these condom users were indisputably at risk for unplanned pregnancy because they did not use other effective contraceptives with the male condom.

In this study sample, 842 (65.0%) women had never tried a COC pill. The main reasons were fear of side-effects (72.1%), satisfied with their current contraceptive (32.1%), and pill-taking was inconvenient for them (18.5%) [Table 2]. Among 453 women who had tried a COC pill, the median age they started use was 24 years (interquartile range, 20-28 years). Use of COC pill was associated with older age (mean ± standard deviation: users and non-users was 33.4 ± 5.8 and 32.4 ± 5.5 years, respectively; t test, P=0.003), not planning to get pregnant (P=0.002), and university education (P=0.004). There was no association with relationship status (P=0.968) or parity (p=0.427). These women preferred the COC pill because of convenience (47.7%), effectiveness (44.8%), menstrual regulation (33.6%), recommendation by their doctor (24.7%), reduced burden to partner (17.0%), for relief of dysmenorrhoea (14.1%), and improvement of acne (12.6%) [Table 3]. They chose a COC pill based on the dosage of hormones, type of hormones, and price. Among the 453 ever-users, 177 (39.1%) had been taking a COC pill in the previous 12 months. Use had stopped in 276 because they feared side-effects (39.1%); they experienced side-effects such as nausea, vomiting, breast tenderness, oedema, or weight gain (27.9%); they consistently forgot to take pills (19.9%); their doctor told them to stop (14.1%); or they were having less frequent coitus (12.3%) (Table 4).

The pattern of contraceptive use in this study sample was similar to that in the 2012 territory-wide survey.1 Male condom was the most popular contraceptive, used by 76.1% of couples in our study. The proportion of women using a COC pill in our study was also similar to that in the 2012 survey. Our survey has provided some information about the characteristics of women who chose to take the COC pill, such as older age, university education, and no plan for future pregnancy. A similar age profile and education attainment were identified in a national survey conducted in the US,6 in which parity and relationship status were also characteristics associated with COC pill use.

Fear of side-effects was the major reason cited by both subgroups of women who stopped or had never tried a COC pill. Studies carried out in both developed and developing countries have also shown that the experience of side-effects as well as the fear of side-effects are major reasons for discontinuation.7 8 9 10 It appeared that fear of side-effects was a unique barrier across different countries and cultures. Minor side-effects such as breast tenderness, fluid retention, nausea, and vomiting were transient and usually subsided after one to two cycles. Major health hazards such as myocardial infarction, stroke, thromboembolism, breast cancer, and cervical cancer are rare. Two meta-analyses showed a 2-fold increase in myocardial infarction and stroke in low-dose COC pill users compared with non-users.11 12 The risk of venous thromboembolism was increased by 3- to 5-fold depending on the type of progestogen used.13 Since the baseline incidence of these vascular events in women of reproductive age is very low (myocardial infarction: 0.2 per 100 000 at age 30-34 years to 2.0 at age 40-44 years14; stroke: 1 per 100 000 at age 30-34 years to 1.6 at age 40-44 years14; thromboembolism: 2 per 10 000 women at reproductive age15), the absolute risk of such vascular complications is very small. Breast cancer risk with a low-dose COC pill is also small. A large meta-analysis of case-control studies from 25 countries showed a modest increase in breast cancer risk with the COC pill (relative risk=1.24; 95% confidence interval [CI], 1.15-1.33).16 The risk of cervical cancer depends on the duration of use. Women who used a COC pill for less than 5 years have no increased risk of cervical cancer. The odds ratio for cervical cancer after using COC for 5 to 9 years was 2.82 (95% CI, 1.46-5.42) and 4.03 (95% CI, 2.09-8.02) for 10 years or longer.17 Cervical cancer is largely preventable by regular cervical smears, safe sex, as well as avoidance of smoking. The overall morbidity and mortality associated with the low-dose COC pill are low and most healthy women can use it without major concerns.

The lack of access to consultation services has exacerbated concern about side-effects, both for women who experience them and for those who fear them.9 At our clinics, women are counselled about the common side-effects and complications of the COC pill. This prevents them from panicking when minor side-effects occur. They are also told to stop taking the COC pill immediately and consult a doctor if they develop signs and symptoms of a major complication. Such counselling helps women establish realistic expectations and they are able to use COC safely. An information sheet detailing side-effects and complications, warning signs and symptoms for major complications, commonly used drugs that interact with COC pill, and missed pill management is given to all users. When first prescribed, we usually provide two packs and then review acceptability after 2 months. Women are advised that different COC pills vary slightly in their side-effect profile and they can change to another formulation if they have problems. We also offer walk-in clinics for any woman who wishes to get contraceptive advice from nurses. The above COC pill delivery mode conforms to the World Health Organization recommendations.18

Apart from side-effects and complications, women should be informed of the non-contraceptive benefits of the COC pill, such as menstrual regulation and relief of dysmenorrhoea; reduced risk for endometrial, ovarian, and colorectal cancers; lower incidence of gynaecological diseases such as endometriosis, pelvic inflammatory disease, ectopic pregnancy; and improved acne and bone health. All such information should be shared with women to help them establish an impartial perspective on the risks and benefits of the COC pill.

The main limitation of this survey is the very low response rate, albeit not unexpected with online survey. There was also selection bias as members of an exclusive group were invited to participate. Those who participated in the survey prompted self-selection bias as they might be systematically different from those who chose not to respond. When we planned the study, we had explored other alternatives such as face-to-face interview, phone interview, or online survey for the general public. Nonetheless, the first would be too expensive and in the last two alternatives, we would be unable to verify respondent’s age or gender. We settled with this arrangement as it was the most convenient means to reach our target group since ESDlife only allowed female members aged 18 to 45 years to participate. The demographic statistics provided by ESDlife revealed that the education attainment and income reported by its female members were better than the population average. The contraceptive choice in this group matched that of the population study and the sample size was not small. Although the results obtained cannot be generalised to the local population, we believe they provide useful insight into the reasons why women do or do not use the COC pill. The other limitation is the number of questions we could ask was limited by the budget. If we had a larger budget to include more questions, we would have explored the type of COC pill used, total duration of use, and the switch pattern in women who used more than one contraceptive in the previous 12 months. Lastly, there was a discrepancy in the number of women who were using a COC pill in the past 12 months. For “Question 15. Are you still on COC in the past 12 months?”, 177 responded positively. In response to Question 7, however, only 163 chose COC pill as one of the contraceptives they had used in the past 12 months. Some women might have omitted COC when they selected their contraceptives from the list provided in Question 7.

The COC pill remains underutilised in Hong Kong compared with many western countries. The male condom is the most popular contraceptive and the proportion of women using a COC pill is one sixth of that of women who use a male condom. A considerable proportion of respondents expressed concerns about actual or anticipated side-effects. Doctors should focus on this area during contraceptive counselling and help dispel the underlying myths and misconceptions surrounding COC pill use. Studies have shown that minor side-effects are transient, major complications are rare in healthy women, and there are many non-contraceptive benefits of the COC pill. These facts should be emphasised during COC counselling to help women balance the risks and benefits of the COC pill and make an informed choice about contraception.

Sponsorship was provided by Pfizer Corporation Hong Kong Limited to cover all costs incurred with ESDlife and incentives for participants. The company was not involved in the study design, execution, data interpretation, or manuscript preparation.

1. Family Planning Knowledge, Attitude and Practice in Hong Kong Survey 2012. Available from: http://www.famplan.org.hk/fpahk/en/template1.asp?style=template1.asp&content=info/research.asp. Accessed 6 Aug 2015.

2. Trussell J. Contraceptive failure in the United States. Contraception 2011;83:397-404. Crossref

3. United Nations. Department of Economic and Social Affairs. Population Division. World contraceptive patterns 2013. Available from: http://www.un.org/en/development/desa/population/publications/pdf/family/worldContraceptivePatternsWallChart2013.pdf. Accessed 6 Aug 2015.

4. Heise L. Beyond acceptability: reorienting research on contraceptive choice. In: Ravindran TS, Berer M, Cottingham J, editors. Beyond acceptability: users’ perspectives on contraception. London: Reproductive Health Matters; 1997: 6-14.

5. User preferences for contraceptive methods in India, Korea, the Philippines, and Turkey. World Health Organization Task Force on Psychosocial Research in Family Planning and Task Force on Service Research in Family Planning. Stud Fam Plann 1980;11:267-73.

6. Hall KS, Trussell J. Types of combined oral contraceptives used by US women. Contraception 2012;86:659-65. Crossref

7. Ali M, Cleland J. Contraceptive discontinuation in six developing countries: a cause-specific analysis. Int Fam Plan Perspect 1995;21:92-7. Crossref

8. Sanders SA, Graham CA, Bass JL, Bancroft J. A prospective study of the effects of oral contraceptives on sexuality and well-being and their relationship to discontinuation. Contraception 2001;64:51-8. Crossref

9. D’Antona Ade O, Chelekis JA, D’Antona MF, Siqueira AD. Contraceptive discontinuation and non-use in Santarém, Brazilian Amazon. Cad Saude Publica 2009;25:2021-32. Crossref

10. Larsson G, Blohm F, Sundell G, Andersch B, Milsom I. A longitudinal study of birth control and pregnancy outcome among women in a Swedish population. Contraception 1997;56:9-16. Crossref

11. Baillargeon JP, McClish DK, Essah PA, Nestler JE. Association between the current use of low-dose oral contraceptives and cardiovascular arterial disease: a meta-analysis. J Clin Endocrinol Metab 2005;90:3863-70. Crossref

12. Khader YS, Rice J, John L, Abueita O. Oral contraceptives use and the risk of myocardial infarction: a meta-analysis. Contraception 2003;68:11-7. Crossref

13. Faculty of Sexual and Reproductive Healthcare Statement. Venous thromboembolism and hormonal contraception. November 2014. Available from: http://www.fsrh.org/pdfs/FSRHStatementVTEandHormonalContraception.pdf. Accessed 6 Aug 2015.

14. Farley TM, Meirik O, Collins J. Cardiovascular disease and combined oral contraceptives: reviewing the evidence and balancing risks. Human Reprod Update 1999;5:721-35. Crossref

15. European Medicines Agency. Benefits of combined hormonal contraceptives (CHCs) continue to outweigh risks. Product information updated to help women make informed decisions about their choice of contraception. London, UK: EMA, 2014. Available from: http://www.ema.europa.eu/ema/index.jsp?curl=pages/medicines/human/referrals/Combined_hormonal_contraceptives/human_referral_prac_000016.jsp&mid=WC0b01ac05805c516f. Accessed 6 Aug 2015.

16. Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53 297 women with breast cancer and 100 239 women without breast cancer from 54 epidemiological studies. Lancet 1996;347:1713-27. Crossref

17. Moreno V, Bosch FX, Muñoz N, et al. Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control study. Lancet 2002;359:1085-92. Crossref

18. World Health Organization. Family Planning: A Global Handbook for Providers. 2011 edition. Available from: https://www.fphandbook.org/sites/default/files/chap_1_eng.pdf. Accessed 6 Aug 2015.

In Hong Kong, thousands of patients undergo breast surgery every year for both benign and malignant conditions.1 Patients expect a good cosmetic outcome and satisfactory postoperative wound management. This is in addition to the expectation of a cure, or in the case of breast cancer, complete removal of lesions with optimal survival.

Several studies have demonstrated that octylcyanoacrylate (OCA) tissue adhesive provides an equivalent cosmetic outcome to wound suturing in repair of lacerations,2 3 head and neck surgery,4 plastic surgery,5 and breast surgery.6 7 The OCA tissue adhesives are supplied as monomers in a liquid form. They polymerise on contact with tissue anions, forming a strong bond that holds the opposed wound edges together. The OCA tissue adhesive usually sloughs off with time. The wound epithelises within 5 to 10 days and the adhesive does not require removal.

In-vitro studies have shown that OCA provides an effective antimicrobial barrier for the first 72 hours after application.8 It is approved by the US Food and Drug Administration as a topical skin adhesive that protects the wound from bacteria. It also facilitates postoperative wound care as patients are allowed to shower immediately. There is no need for suturing, staple removal, or dressings. Higher patient satisfaction following skin closure with OCA tissue adhesive compared with sutures has been observed.6 Studies also show faster wound closure with OCA.9

This study aimed to assess the outcome of elective breast surgical incision repair with OCA tissue adhesive compared with standard wound suture (SWS). We compared the cosmetic outcome, complication rates, and patient satisfaction score for breast incisions in elective surgery closed with OCA tissue adhesive versus SWS.

The study was in compliance with the Declaration of Helsinki and ICH-GCP (International Conference on Harmonisation, Good Clinical Practice). It was reviewed and approved by the institutional review boards.

Based on a randomised trial of OCA versus SWS in breast surgery,5 patient satisfaction in an OCA group has been reported to be significantly higher than that of a SWS group. To detect a difference with a power of 90% and α=0.05, 35 patients were needed for each arm.

A total of 70 female patients, who underwent elective excision of clinically benign breast lump between February 2009 and November 2011 in this randomised controlled trial, were randomly allocated to have wound closure using either OCA or SWS with a continuous monofilament subcuticular method. They were seen on the morning of the surgery, consented, and randomly allocated to a study arm. Each randomisation number was computer-generated, sealed in an envelope, and kept in a secure designated place. At the time of wound closure, the surgeon would call a third-party nurse to open the sealed envelope that would determine the method to be used for wound closure. The surgery was performed by specialists or surgical trainees under a specialist’s supervision. Two evaluation forms were administered to collect information on wound condition and cosmetic grading by different parties.

Postoperatively, the wound condition was examined by a surgeon who was not involved in the study. An evaluation form was completed to note any indication of (1) wound infection, (2) dehiscence, (3) oozing, and (4) discharge on day 0-1 (early postoperative period) and day 10-14 (first follow-up).

The cosmetic grading of the surgical wound was checked on day 30 and 180 by an evaluator (surgeons not involved in the study or the above evaluations) who looked for any sign of (1) step-off borders (edges not on same plane), (2) contour irregularities (wrinkled skin near wound), (3) margin separation (gap between sides), (4) edge inversion (wound not properly everted), and (5) excessive distortion (swelling/oedema/infection); and (6) evaluated the overall appearance of the wound.

Patient evaluation of whether the appearance of the wound was “good” or “bad” over a score of 1 (very bad) to 10 (very good) on day 30 and day 180 was also recorded.

Patients were also asked five questions about self-care of the wound at the day-30 visit. Questions were answered and rated on a 5-point Likert scale (very good, good, neutral, bad, very bad) regarding (1) pain, (2) ease of caring, (3) self-showering, (4) frequency of hospital/clinic visits for wound cleansing, and (5) comfort level of wound dressing.

Data were summarised with descriptive statistics. Means and standard deviations (for numeric variables) and numbers and percentages (for categorical variables) were calculated where appropriate. We checked the normality of the data and found that it did not follow the normal distribution. Therefore the Wilcoxon rank sum test and Fisher’s exact test were applied to determine any significant difference between the OCA and SWS groups. All statistical analyses were done using the Statistical Package for the Social Sciences (Windows version 16.0; SPSS Inc, Chicago [IL], US) and R version 3.0.2 (the R Foundation). All statistical tests were two-sided and statistical significance was considered at P<0.05.

A total of 70 patients, half of whom were randomised to receive OCA or SWS, were entered into this study. One patient from the suture group was lost to follow-up and excluded from subsequent analysis, leaving a total number of 69 patients (35 for OCA group and 34 for SWS group).

The demographics of the two groups were comparable. There was no statistical difference in terms of age, tumour size, co-morbidity including diabetes, pathology, laterality and location of the lesion, or the rank of the surgeon involved (Table 1).

Use of OCA was associated with significantly less time to complete the wound closure process compared with suture (mean, 80.6 seconds vs 344.6 seconds; P<0.001; Table 1). With similar length of surgical wound, OCA required 7.1 times less time to close the wound than suture (P<0.001).

The occurrence of adverse wound condition is shown in Table 2. There was no unfavourable condition noticed upon first-day follow-up. Wound complications on day 10-14 all occurred in the OCA group.

Table 3 summarises the incidence of any wound cosmetic problem on day 30 and 180. Cosmetic problems were found only on day 30, and were not confined to any one group. One patient from the suture group felt that the overall appearance of the surgical wound on day 30 was bad but subsequently commented it was “good” on day 180. No bad comments were received from any patient who had undergone wound closure with OCA.

For the patient’s view of cosmetic outcome, a higher score was given to wounds closed by sutures compared with OCA on both day 30 and 180, although the standard deviations were larger in the OCA group, and the differences were not statistically significant (Table 4). Higher scores were given on day 180 compared with day 30 in both groups.

The actual number of hospital or clinic visits was recorded. Patients in the OCA group required fewer visits than those in the suture group (16 OCA vs 18 sutures, 1.19 ± 2.66 vs 2.50 ± 4.57; P=0.063). A higher percentage of patients in the OCA group felt “very good” on ‘self-showering’ (OCA vs suture, 66.7% vs 21.2%; P<0.001), ‘frequency of hospital/clinic visits for wound cleansing’ (OCA vs suture, 66.7% vs 25.0%; P=0.001), and ‘comfort level of wound dressing’ (OCA vs suture, 58.8% vs 18.2%; P=0.003) [Table 5]. More patients in the suture group rated “good”, instead of “very good” for these three categories. The same applied for ‘ease of caring’ although statistical significance was not reached. For patients who commented “bad” or “very bad”, the percentages were generally higher from the OCA group. There was no significant difference between the two groups for reports about pain (P=0.564).

Tissue adhesive material has long been used in wound closure in western countries, and offers the advantages of faster closure, need for less postoperative wound care, and higher patient satisfaction.

In this study among Chinese women, we demonstrated that time required for wound closure was much less in the OCA group compared with the SWS group (P<0.001). Although the difference was significant, time required for wound closure was not a significant concern for the surgeon.

There were three instances of postoperative complications in the OCA group. In two patients, the surgery was performed by a trainee under supervision, and in one by a specialist. One patient required secondary suturing 3 weeks later. The other two were treated conservatively with antibiotics. We postulate that there is a learning curve for closure with OCA, thus technical skill and experience of the surgeon may play a role.

For cosmetic outcome, the score was comparable in both groups, although slightly higher in the SWS group. Nonetheless, the difference was less than 1 point on a scale of 10. The standard deviation in the OCA group was wider (OCA 2.17 vs SWS 1.36) at day 30 but was not statistically significant.

For wound problems, margin separation occurred on postoperative day 30 in five patients in the OCA group compared with two in the SWS group. There was no statistical difference in cosmetic problem grading between the two study groups. It should be noted that tissue adhesive wound repair is a manual skill, just like suturing, and requires practice and careful application. Factors such as wound oozing or discharge may hinder proper functioning of the tissue adhesive.

The skin of patients of Chinese or Asian descent is more prone to keloid scarring and pigmentation, thus the effect of using tissue adhesives may differ to that of a western population. A previous study did not show any difference in the rate of hypertrophic scar formation.10 In our study, there was no hypertrophic scar or keloid formation in either group.

There was a significant difference in preference in terms of self-showering, frequency of hospital/clinic visits, and comfort level of dressing between OCA and SWS groups. These factors affect patients since they impact on daily activities and saving of time. On the other hand, there was no statistical difference in degree of pain, although the Phi value was very small, thus a larger sample size may be required to detect any difference. The sample size for all other values was adequate.

In terms of cost, a study in Hong Kong has shown that tissue adhesive is more expensive than SWS,11 but may be more cost-effective from a social viewpoint in terms of superior cosmetic outcome and overall patient satisfaction.

First, this was a prospective randomised study confined to closure of benign breast lump wounds and thus not necessarily applicable to all surgical wounds. We did not determine the number of eligible patients for the study, or note how many of them refused to take part. This may have led to self-selection bias. Second, the operation was performed by different surgeons of different seniority, and may have led to varying levels of surgical technique. This was not shown to be statistically significant but the sample size was not designed to reflect this. Lastly, during wound evaluation, the surgeon might not have been totally blind to the type of closure if sutures or stitch marks were visible. This may be a cause of blinding bias. The use of a scoring scale is also a subjective evaluation, and may be influenced by the patient’s mood and other factors.

The use of OCA should be discussed and offered to patients as an option for wound closure in breast surgery. It achieves faster wound closure, is not inferior to standard wound closure, has a higher comfort level, and requires less frequent clinic visits. Understanding cost-effectiveness is essential in medical care, thus OCA should be offered as an option to be provided as a self-financed item in the public sector where it is now widely available.

We thank Mr Jack Chau and Miss Fidelia Wong for performing the early statistical analysis, and Mr Ling-hiu Fung and Mr Wing-pan Luk of the Hong Kong Sanatorium & Hospital for the later review and re-running of the statistical analysis methods. We also thank Ethicon, Johnson and Johnson Company for providing the OCA study material.

Johnson and Johnson provided the study material. No other conflict of interests was declared by authors.

1. Top ten cancers in 2013. Hong Kong Cancer Registry. Available from: http://www3.ha.org.hk/cancereg/statistics.html. Accessed Jan 2016.

2. Singer AJ, Quinn JV, Clark RE, Hollander JE; TraumaSeal Study Group. Closure of lacerations and incisions with octylcyanoacrylate: a multicenter randomized controlled trial. Surgery 2002;131:270-6. Crossref

3. Chow A, Marshall H, Zacharakis E, Paraskeva P, Purkayastha S. Use of tissue glue for surgical incision closure: a systematic review and meta-analysis of randomized controlled trials. J Am Coll Surg 2010;211:114-25. Crossref

4. Maw JL, Quinn JV, Wells GA, et al. A prospective comparison of octylcyanoacrylate tissue adhesive and suture for the closure of head and neck incisions. J Otolaryngol 1997;26:26-30.

5. Toriumi DM, O’Grady K, Desai D, Bagal A. Use of octyl-2-cyanoacrylate for skin closure in facial plastic surgery. Plast Reconstr Surg 1998;102:2209-19. Crossref

6. Gennari R, Rotmensz N, Ballardini B, et al. A prospective, randomized, controlled clinical trial of tissue adhesive (2-octylcyanoacrylate) versus standard wound closure in breast surgery. Surgery 2004;136:593-9. Crossref

7. Nipshagen MD, Hage JJ, Beekman WH. Use of 2-octylcyanoacrylate skin adhesive (Dermabond) for wound closure following reduction mammaplasty: a prospective, randomized intervention study. Plast Reconstr Surg 2008;122:10-8. Crossref

8. Quinn JV, Osmond MH, Yurack JA, Moir PJ. N-2-butylcyanoacrylate: risk of bacterial contamination with an appraisal of its antimicrobial effects. J Emerg Med 1995;13:581-5. Crossref

9. Soni A, Narula R, Kumar A, Parmar M, Sahore M, Chandel M. Comparing cyanoacrylate tissue adhesive and conventional subcuticular skin sutures for maxillofacial incisions—a prospective randomized trial considering closure time, wound morbidity, and cosmetic outcome. J Oral Maxillofac Surg 2013;71:2152.e1-8. Crossref

10. Wilson AD, Mercer N. Dermabond tissue adhesive versus Steri-Strips in unilateral cleft lip repair: an audit of infection and hypertrophic scar rates. Cleft Palate Craniofac J 2008;45:614-9. Crossref

11. Wong EM, Rainer TH, Ng YC, Chan MS, Lopez V. Cost-effectiveness of Dermabond versus sutures for lacerated wound closure: a randomised controlled trial. Hong Kong Med J 2011;17 Suppl 6:4-8.

Breast cancer is the second leading cause of death due to cancer in the world with an age-standardised incidence rate of 43.1 per 100 000 population in 2012.1 Gøtzsche and Nielsen2 showed that early detection of breast cancer can reduce mortality. The benefit of mammographic screening in terms of lives saved is greater than the harm caused by overdiagnosis; according to Duffy et al,3 2 to 2.5 lives are saved for every overdiagnosed case.

Breast cancer is also a significant health problem in Hong Kong. It is the third leading cause of death due to cancer and the most common cancer of women.4 The crude incidence rate of breast cancer in Hong Kong increased from 57 per 100 000 in 2000 to 91.7 per 100 000 in 2012.4

Despite this, Hong Kong does not have a universal breast cancer screening programme for the whole population. Women who wish to be screened must arrange and pay for it. As the incidence of breast cancer in Hong Kong is low compared with western populations, there is concern about the cost-effectiveness of a universal screening programme. The incremental cost-effectiveness ratio for mammography examination is relatively higher than in the United States.5

In view of the controversy there is a need for further studies in Hong Kong to provide local data on the efficacy of breast cancer screening by mammogram. This will enable policy makers, doctors, and patients to decide on the most cost-effective method of early breast cancer detection.

This study aimed to investigate whether there are any prognostic differences between screen-detected (mammography, ultrasound examination of breasts, or clinical examination) and self-detected breast cancers in a cohort of Hong Kong breast cancer patients and to determine whether there is any benefit of detection by screening.

A retrospective study was conducted at the Hong Kong Sanatorium & Hospital (HKSH), a private hospital in Hong Kong where the Breast Care Centre provides a comprehensive breast screening programme and breast cancer consultation services for patients. Since 2003, all breast cancer cases in the hospital have been discussed at a weekly multidisciplinary breast conference. Over 50% of cases are referred from public hospitals spread across the whole territory.

Clinical history, and information about diagnosis, neoadjuvant chemotherapy, surgical treatment, postoperative pathology, and treatment recommendations for each patient are recorded in a structured datasheet before the conference. The Chairman of the conference validates data by checking the data logic during case presentation and, if necessary, clarifying details with the doctor-in-charge. A research assistant again checked data validity and logic by computer for cases selected for analysis. Frequency tables, scatter plots, and cross-tabulation tables were generated for each required variable to ensure completeness and to determine whether any data deviated from usual clinical practice.

All Chinese females confirmed to have in-situ or invasive breast cancer from or referred to HKSH between 2003 and 2010 were included in this study. For analysis of trends of prognostic factors, only patients between 2004 and 2010 were included as the sample size for year 2003 was small after dividing data into subsets by year and prognostic factors.

The use of the database for data analysis for health care research purposes was approved by the Research Ethics Committee of HKSH.

Patients were classified into two groups, screen-detected or self-detected tumour, before the outcomes were reviewed. The screen-detected group included screening mammogram, screening ultrasonogram, or clinical examination. The self-detected group included self-examination or presence of symptoms at presentation. Such information was recorded on the datasheet that was anonymised. Demographic data of patients were retrieved and significant prognostic factors according to St Gallen’s risk categorisation were analysed.

The Statistical Package for the Social Sciences (Windows version 18.0; SPSS Inc, Chicago [IL], US) was used for analysis. Descriptive statistics were used to summarise demographics, as well as pathological and clinical characteristics of patients. Univariate logistic regression was used to determine odds ratio (OR) of screening status for different pathological characteristics. Association between type of surgery and screening status, tumour size, and age was determined by univariate logistic regression. In order to know how effective the models were in predicting the type of surgery, Hosmer-Lemeshow goodness-of-fit test was used, in which the null hypothesis was no significant difference between observed and predicted values of dependent variable. Multivariate analysis was performed on the type of surgery, which was dichotomous having breast-conserving surgery (BCS) as reference group with adjustment of confounding factors that included detection mode, lymph node status, tumour size, tumour grade, tumour stage, oestrogen receptors (ER), progesterone receptors (PR), HER2 score, and age. Nagelkerke’s R² was used to show explanatory power of model.

Data for invasive tumour size were plotted against screening status to gain an overview of changes between 2004 and 2010.

Data from 2763 Chinese women out of 3373 cases confirmed to have in-situ or invasive breast cancer between 2003 and 2010 were analysed. Of the 610 cases excluded from analysis, 210 were due to unknown histology, 258 with unknown report type, and 142 with unknown first-detection method. Of the patients included, 75.7% were in the self-detected group and 24.3% in the screen-detected group. The mean age of patients was 50.2 (range, 24-92) years, with the highest number in both groups aged 40 to 49 years (Table 1). There was also a significant proportion of younger patients (<40 years) in the self-detected group (16.6%). The screen-detected group had a statistically significant higher age at first live birth, though the difference in mean age was only 1 year.

Pathological characteristics of self-detected and screen-detected groups are shown in Table 1. There was a significant difference in ER, but no demonstrable significant difference in PR or HER2 status. Approximately 4% of patients in both groups did not undergo surgery for a variety of reasons, thus only limited information was available from biopsy specimens about pathological characteristics.

The odds of having 0 lymph nodes, smaller tumour size, or ER/PR positivity were all higher in the screen-detected group. Results were not statistically significant for HER2 positivity although it showed higher odds in the self-detected group (Table 2).

Table 3 shows that after adjustment for potential confounding factors, patients with screen-detected cancers were less likely to require mastectomy (OR=0.658, P=0.004). Statistically significant factors associated with a higher risk of mastectomy included: positive lymph node, tumour grading higher than 1, tumour size of >2 cm, older age, and positive HER2 score. The Hosmer-Lemeshow test had a P value of 0.88, meaning the goodness of fit of the model was satisfactory at the 5% significance level.

The Figure shows the trend in differences between groups for invasive tumour size between 2004 and 2010. In the screen-detected group, there was an increasing proportion of invasive tumours detected when ≤2 cm and a decreasing trend for detection of tumour of >2 to 5 cm. There was, however, no significant difference between groups for trend of histology, lymph node involvement, tumour grade, type of surgery, ER positivity, or HER2 positivity.

Although the incidence of breast cancer in Hong Kong is half that of the United Kingdom, the screen-detected group showed a pattern of breast cancer diagnosis at an earlier stage compared with the self-detected group. This is consistent with the findings of similar studies in other countries, such as Singapore.6 This study may provide evidence that supports the benefits of regular screening to detect breast cancer lesions at an early stage in Hong Kong women. This will facilitate less invasive surgery and a possibly better overall clinical outcome. Breast cancer screening programmes have been established in many countries around the world. In order to bring Hong Kong in line with world health care standards, more research that will result in established and unified guidelines for the local population is required.

Significant prognostic factors in the St Gallen’s risk categorisation, including number of lymph nodes with disease, size of primary tumour, and histological grade between the self-detected and screen-detected groups were analysed. These prognostic factors were chosen because they have been verified as significant in the local population.7

In this study, screen-detected breast tumours were of smaller size, at a lower stage and grade, and with less lymph node involvement. Screen-detected breast cancers in the Hong Kong population may thus carry a better prognosis than self-detected ones. This can serve as evidence that fulfils most of the Wilson and Jungner criteria for a screening evaluation programme8 and also supports breast cancer screening in Hong Kong. Some of the criteria for breast cancer screening that have been fulfilled include: an important health problem (it is the second leading cause of death from cancer in Hong Kong), acceptable treatment is well established, facilities for diagnosis and treatment are widely available, natural history of the disease is well understood, and effective treatment is available for early-stage disease. Findings of this study suggested that breast tumours detected on screening have a better prognosis. The cost-benefit balance was not addressed, however, nor screening/case-finding policies.

In the screen-detected group, a higher percentage of tumours were ER positive. It was revealed that ER and PR are significant prognostic factors within the first 10 years following diagnosis.9 It is known that HER2 positivity shows poorer prognosis8 but there was no significant difference in HER2 status between screen-detected and self-detected groups.

By observing the trend in size of invasive tumour at first presentation, the stable pattern in the self-detected group suggests that tumour detection by the general population has not improved. On the contrary, an increasing detection of tumours of ≤2 cm in the screen-detected group is an evidence of the improved efficacy of screening using new technology such as mammogram or sonogram. There remains room for improvement in the application of radiology. Apart from magnetic resonance imaging, digital mammography may be more efficacious in women younger than 50 years.9 This may change future trends in early diagnosis.

Unlike tumour size and number of lymph nodes involved, both of which are increased in breast tumours detected at a later stage with a consequent poorer prognosis, prognostic factors such as status of ER/PR and HER2 are intrinsic characteristics of tumours. They should not differ whether or not a tumour is detected at an earlier stage. Therefore no statistically significant difference in these intrinsic characteristics was expected between screen-detected and self-detected tumours. Nonetheless, in this study, tumours in the screen-detected group were more likely to be positive for ER, and this may carry some prognostic implication. Further studies may be required to investigate whether tumours detected at an earlier stage show differences in intrinsic factors.

When choosing between BCS and mastectomy, detection mode, number of positive lymph nodes, invasive tumour size, grading, staging, ER, PR, HER2 score, and age were potential significant factors. With adjustment of these factors, detection mode may be an independent factor that affects choice of surgery. Screen-detected patients tended to have BCS when the effect of number of positive lymph nodes, tumour size, grading, HER2 score, and age was excluded. Such surgery is less invasive than mastectomy and is associated with better cosmetic outcome, and may have an important impact on the psychological health and coping ability of patients recovering from breast cancer. Breast cancer screening may lead to less invasive treatment with better rehabilitation outcome.

Many studies have claimed longer survival in patients with breast cancer detected by screening mammogram. Nonetheless, this may be due to lead-time bias: survival time appears longer because diagnosis is earlier than in patients where tumour has been self-detected or become symptomatic. In addition, there is selection bias since women with a family history of breast cancer or who are better informed are more likely to submit to breast cancer screening.

A local study suggested that population-based breast cancer screening by mammography may not be cost-effective in Hong Kong women.5 This balance between cost and benefit may be altered by the rising incidence of breast cancer in Hong Kong and the availability of advanced breast imaging technology that is associated with fewer false-positive diagnoses. There may also be a broader range of screening options, hence cost will be lowered. Therefore, the incremental cost-effectiveness ratio may be lowered.

A prospective randomised controlled trial would be the most effective study design to evaluate the effectiveness of mammogram screening. This would require a huge amount of resources, however. In addition, education level and household income, which are risk factors for breast cancer, may affect a woman’s decision to undergo a mammogram. These confounding factors should be considered when determining the effect of mammogram examination on development of breast cancer. Postmenopausal hormone replacement therapy also affects the density of breast tissue that may hinder the effectiveness of mammography for breast cancer screening.10

The lack of a population-based breast cancer screening programme in Hong Kong should prompt study of the attitude of Hong Kong women towards breast cancer screening. Quantitative surveys or qualitative interviews such as focus groups could help determine their opinion of mammogram screening, what proportion of women perform regular self-examination or undergo clinical breast examinations and regular mammogram examinations and how often, and whether age is a contributing factor. Trend study may also be meaningful if a particular age-group shows an increasing or decreasing trend for any of the examinations. Understanding level of knowledge about risk factors for breast cancer can also guide appropriate education about breast cancer prevention.

The major strength of this study is the large number of cases in the database, which is one of the most comprehensive breast cancer databases available in Hong Kong in terms of surgical and pathological characteristics. This provides valuable information about the characteristics of breast cancers detected by oneself and through screening, thus allowing a better understanding of the potential benefits of screening by mammogram or ultrasound examination.

This study has limitations. First, it was not a randomised controlled trial. The presence of confounding factors such as living standard, household income, and education level could not be totally excluded. There were also more self-detected than screen-detected patients in this study, thus data might skew towards self-detected cases. It is ethically difficult to randomise women to a control group of education and regular breast self-examination, or an intervention group of regular breast cancer screening by imaging.

Second, data were derived from a single private hospital and findings may not be representative of the Hong Kong population in general. Self-selection bias, especially for attending a private hospital, is also possible. Nonetheless, this is probably one of the largest breast cancer databases in Hong Kong, thus one of the best available sources of information options at present.

Third, secondary data that had been used in this study may not be in a format that met the research question. Some information required may not be available from secondary data. Only 2198 (79.6%) patients were included in the multivariate logistic regression. Other medical and non-medical factors that could have affected the choice of surgery in individual patients might be related to the practice of screening. Data on parity of women and breastfeeding experience, which may be of interest/relevance, were also not available. Also, the potential benefits and harm of screening were not thoroughly examined due to the unavailability of data for survival, mortality, and side-effects. For this study, a long period of time was required to examine the data and filter out required variables for analyses as there were more than 200 variables in the database. Information bias also exists as complete blinding of the analysts was not possible.

This study suggests that in the local Hong Kong population, breast cancers detected by screening mammogram or ultrasound tend to be of smaller size, lower stage, lower grade and with less lymph node involvement, and consequent better prognosis. Although this may not be considered conclusive evidence to support regular screening imaging of Hong Kong women on a population-wide basis, it provides indirect evidence that women in our local population may gain clinical benefit from such a programme.

This article is adapted from a dissertation submitted in partial fulfilment of the requirements for the Master of Public Health at the University of Hong Kong, Hong Kong Special Administrative Region, China. The dissertation was awarded distinction in 2012. Part of the material in this article was presented in the Hong Kong Sanatorium & Hospital Li Shu Pui Symposium 2012 in Hong Kong.

The first author would like to acknowledge Dr Joseph Wu, Assistant Professor, School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, for his supervision of the project leading to completion of the dissertation for the Master of Public Health course. The authors would also like to thank Dr Andrew Ho, Senior Research Assistant, who is statistical advisor of the project. Finally, the authors would like to express sincere appreciation to staff of Information Technology Department of Hong Kong Sanatorium & Hospital, who helped retrieve all necessary data for the study.

All authors have disclosed no conflicts of interest.

1. GLOBOCAN 2012: Estimated cancer incidence, mortality and prevalence worldwide in 2012. Available from: http://globocan.iarc.fr/Pages/fact_sheets_population.aspx. Accessed Dec 2015.

2. Gøtzsche PC, Nielsen M. Screening for breast cancer with mammography. Cochrane Database Syst Rev 2011;(1):CD001877. Crossref

3. Duffy SW, Tabar L, Olsen AH, et al. Absolute numbers of lives saved and overdiagnosis in breast cancer screening, from a randomized trial and from the Breast Screening Programme in England. J Med Screen 2010;17:25-30. Crossref

4. Hong Kong Cancer Registry. Female breast cancer in Hong Kong. Available from: http://www3.ha.org.hk/cancereg/breast_2012.pdf. Accessed Jan 2016.

5. Wong IO, Kuntz KM, Cowling BJ, Lam CL, Leung GM. Cost-effectiveness analysis of mammography screening in Hong Kong Chinese using state-transition Markov modelling. Hong Kong Med J 2010;16(Suppl 3):38S-41S.

6. Chuwa EW, Yeo AW, Koong HN, et al. Early detection of breast cancer through population-based mammographic screening in Asian women: a comparison study between screen-detected and symptomatic breast cancers. Breast J 2009;15:133-9. Crossref

7. Yau TK, Soong IS, Chan K, et al. Evaluation of the prognostic value of 2005 St. Gallen risk categories for operated breast cancers in Hong Kong. Breast 2008;17:58-63. Crossref

8. Wilson JM, Jungner G. Principles and practice of screening for disease. Public Health Papers. No. 34. Geneva: World Health Organization; 1968.

9. Soerjomataram I, Louwman MW, Ribot JG, Roukema JA, Coebergh JW. An overview of prognostic factors for long-term survivors of breast cancer. Breast Cancer Res Treat 2008;107:309-30. Crossref

10. Cox B, Ballard-Barbash R, Broeders M, et al. Recording of hormone therapy and breast density in breast screening programs: summary and recommendations of the International Cancer Screening Network. Breast Cancer Res Treat 2010;124:793-800. Crossref