Hong Kong Med J 2015 Dec;21(6):528–35 | Epub 16 Oct 2015
© Hong Kong Academy of Medicine. CC BY-NC-ND 4.0
Early postoperative outcome of bipolar transurethral enucleation and resection of the prostate
CL Cho, FRCSEd (Urol), FHKAM (Surgery); Clarence LH Leung, MRCSEd; Wayne KW Chan, FRCSEd (Urol); Ringo WH Chu, FRCSEd (Urol), FHKAM (Surgery); IC Law, FRCSEd (Urol), FHKAM (Surgery)
Division of Urology, Department of Surgery, Kwong Wah Hospital, Yaumatei, Hong Kong
Corresponding author: Dr CL Cho (firstname.lastname@example.org)
Click here to watch a video clip showing transurethral enucleation and resection of the prostate technique
Objectives: To report the early postoperative outcome of bipolar transurethral enucleation and resection of the prostate. Our results were compared with those published from various centres.
Design: Case series.
Setting: Regional hospital, Hong Kong.
Patients: A total of 28 consecutive patients who had undergone bipolar transurethral enucleation and resection of the prostate by a single surgeon between January and June 2014. All patients were evaluated preoperatively by physical examination, digital rectal examination, transrectal ultrasonography, and laboratory studies, including measurement of haemoglobin, sodium, and prostate-specific antigen levels. Patients were assessed perioperatively and at 4 weeks and 3 months postoperatively.
Results: The mean resected specimen weight of prostatic adenoma in 28 patients was 48.2 g with a mean enucleation and resection time of 13.6 and 47.7 minutes, respectively. There was a mean decrease in serum prostate-specific antigen by 85.9% (from 6.4 ng/mL to 0.9 ng/mL) postoperatively. Prostate volume was decreased by 68.2% (from 71.9 cm3 to 22.9 cm3) at 4 weeks postoperatively. The mean postoperative haemoglobin drop was 11.5 g/L. The rate of transient urinary incontinence at 3 months was 3.6%. Patients who underwent bipolar transurethral enucleation and resection of the prostate had a short catheterisation time and hospital stay, which is comparable to conventional transurethral resection of the prostate.
Conclusions: Bipolar transurethral enucleation and resection of the prostate should become the endourological equivalent to open adenomectomy with fewer complications and short convalescence. The technique of bipolar transurethral enucleation and resection of the prostate can be acquired safely with a relatively short learning curve.
New knowledge added by this study
- Bipolar transurethral enucleation and resection of the prostate (TUERP) achieves satisfactory early functional outcomes and is associated with low morbidity. The technique is applicable to prostates of all size.
- Outcomes comparable with large case series could be achieved with a short learning curve.
- Bipolar TUERP should be the technique of choice for a large-sized prostate.
- Bipolar TUERP is an alternative to conventional transurethral resection for small and medium-sized prostates.
Despite the availability of numerous minimally invasive techniques, transurethral resection of the prostate (TURP) remains the most common surgical treatment for lower urinary tract symptoms (LUTS) caused by benign prostatic enlargement (BPE) in small to medium-sized prostates.1 Nonetheless, TURP has been associated with significant complication rates.2
Bipolar TURP uses saline irrigation, which decreases the risk of TURP syndrome compared with monopolar TURP, and both bipolar and monopolar TURP result in comparable functional outcomes.3 The bipolar system can also be broadened to enucleate the prostate gland along the surgical capsule, using a resectoscope combined with a loop. This transurethral enucleation and resection of the prostate (TUERP) technique can potentially remove more prostatic tissue than TURP and requires no additional devices.
In the present study, we describe the technique and early postoperative outcomes of bipolar TUERP and compare our results with major international series.
Between January 2014 and June 2014, 28 consecutive patients underwent bipolar TUERP at Kwong Wah Hospital, Hong Kong. All patients were evaluated preoperatively by physical examination, digital rectal examination, transrectal ultrasonography (TRUS) of the prostate, and laboratory studies that included measurement of haemoglobin, sodium, and prostate-specific antigen (PSA). Patients were offered the option of ultrasound-guided transrectal prostate biopsy if the PSA level was >4 ng/mL or if the digital rectal examination showed suspicion of prostate cancer. Abnormal digital rectal examination findings included prostate nodule, asymmetry of the lateral lobes, or irregularity of the prostate. The TRUS was performed to measure the maximum length, width, and anteroposterior height of the prostate to calculate the prostate volume using the ellipse formula, where prostate volume (mL) = 0.52 x length x width x height. Patient baseline characteristics, indications for surgery, and operative data and complications were recorded by doctors. Patients with neurogenic bladder, previous genitourinary tract surgery, urethral stricture, or known bladder or prostate carcinoma were excluded from the technique of bipolar TUERP.
Equipment and technique
All bipolar TUERP procedures were performed by a single surgeon. This surgeon had performed 37 TUERPs using various techniques and devices previously, before the procedure was standardised as described below and shown in Figure 1. The technique used in this report was first described by and adopted from Prof CX Liu at Zhujiang Hospital of Southern Medical University in Guangzhou.4
Figure 1. Operative steps of bipolar transurethral enucleation and resection of the prostate
(a) Incision starts close to verumontanum (V). (b) Distal median lobe (ML) is dissected from surgical capsule (SC). (c and d) Left and right lateral lobes (LL and RL) are detached from surgical capsule by resectoscope sheath and denuded vessels are cauterised. (e) Median lobe is detached from bladder neck (BN) by loop electrode. (f) Subtotally enucleated right lobe is resected rapidly without haemorrhage
Antiplatelet medications were stopped 3 days prior to surgery. Patients received general or spinal anaesthesia and were placed in the lithotomy position. Bladder stones where present were fragmented with a holmium laser via a 21-Fr rigid cystoscope and were evacuated with an Ellik evacuator before bipolar TUERP. A 26-Fr Olympus SurgMaster TURis resectoscope (Olympus Europe, Hamburg, Germany) with a standard loop was used. The incision was begun immediately proximal to the verumontanum using a cutting current. The surgical capsule plane was identified, and the whole gland dissected in a retrograde fashion from the cleavage plane using the resectoscope sheath, until the circular fibres of the bladder neck were identified. The loop electrode was used to coagulate all of the denuded vessels immediately during the detachment process. The adenoma was subtotally enucleated with a narrow pedicle attached to the bladder neck at the 6 o’clock position. The devascularised adenoma was rapidly resected in pieces by the loop electrode. The bladder neck at 5 to 7 o’clock was removed if it appeared relatively high. The anterior commissure at 12 o’clock was preserved except when it appeared obstructive endoscopically. The chips were evacuated with an Ellik evacuator. Finally, the prostatic fossa was inspected and haemostasis secured. A 24-Fr three-way urethral catheter was inserted at the end of the procedure for bladder irrigation. One of the patients in the series had open inguinal hernia repair performed after bipolar TUERP. Haemoglobin level and serum sodium concentration were measured on the same day after surgery. The protocol for postoperative care following bipolar TUERP was the same as that for monopolar and bipolar TURP in our unit. Bladder irrigation was stopped the following morning, and the catheter was removed on the second day postoperatively.
All patients were evaluated following bipolar TUERP during clinic visits at 4 weeks and 3 months. At each visit, history, physical examination, International Prostate Symptom Score (IPSS), and TRUS of the prostate were evaluated. The presence or absence of transient urinary incontinence was documented with direct questioning of the patient. Uroflowmetry was performed at 8 weeks, and serum PSA levels were measured at 3 months.
Table 1 lists the patients’ baseline characteristics, operative data, and early postoperative outcomes. Enucleation time was defined as the time from incision to completion of subtotal enucleation of the adenomatous tissue. Resection time was defined as the time needed for fragmentation of the en-bloc adenoma into chips. The mean enucleation and resection times were 13.6 (median, 15; range, 10-30) minutes and 47.7 (median, 35; range, 15-120) minutes, respectively, with a mean of 48.2 g of adenoma resected.
Table 1. Baseline characteristics, operative data, and early postoperative outcomes in patients with transurethral enucleation and resection of the prostate
The mean PSA level decreased from 6.4 ng/mL to 0.9 ng/mL at 3 months postoperatively, representing an 85.9% decrease. Pathological examination of enucleated tissue revealed prostatic adenocarcinoma in one patient who had T1a disease with a Gleason score of 6; the serum PSA level decreased from 4.6 ng/mL to 1.7 ng/mL in this patient. There was a significant decrease in mean TRUS volume from 71.9 cm3 to 22.9 cm3 at 4 weeks and to 15.1 cm3 at 3 months postoperatively, corresponding to decreases of 68.2% and 79.0% at 4 weeks and 3 months, respectively.
More than half of the patients in our series (16 of 28 patients) presented with refractory acute urinary retention or obstructive uropathy and had required catheterisation prior to surgery. Preoperative uroflowmetry within the last year was available in only 15 patients, thus comparison between preoperative and postoperative urodynamic parameters was less representative. The mean peak urinary flow rate was 20.9 mL/s, and the mean post-void residual was 31.6 mL at 8 weeks postoperatively. The mean IPSS was 9.4, and the mean quality-of-life score was 1.9 at 4 weeks.
There was no requirement for blood transfusion nor incidence of clot retention in any patient. The mean decrease in haemoglobin was 11.5 g/L. Urinary tract infection presenting as acute epididymitis was noted in two (7.1%) patients. One (3.6%) patient required re-catheterisation on day 2 postoperatively and was successfully weaned off the catheter on day 5. Transient urinary incontinence was noted in three patients and one patient at 1 and 3 months postoperatively, respectively (10.7% at 1 month and 3.6% at 3 months). An average of two incontinence pads were required daily, and all cases of transient urinary incontinence subsided within 4 months. No urethral stricture, meatal stenosis, or bladder neck contracture was noted at 3 months.
The TURP has been considered the standard surgical therapy for LUTS caused by BPE. Despite improvements in equipment and techniques over the years, TURP remains associated with significant morbidity and re-treatment rates, particularly in patients with a large prostate.5 Open prostatectomy (OP) is therefore still considered a valid option for patients with a prostate of >80 g.6
Surgical enucleation for the treatment of LUTS caused by BPE remains the most complete method to remove adenomas of any size; the history of surgical enucleation dates back more than 100 years.7 In spite of the low re-operation rate and high success rate, OP is an invasive procedure associated with higher transfusion rates, longer catheterisation time, and longer hospital stay. As a result, the popularity of OP has declined.
The concept of surgical enucleation was revisited with the advent of endoscopic alternatives to open enucleation. Endoscopic enucleation allows for maximal removal of the adenoma and results in potentially equivalent efficacy compared with its open counterpart, with significantly lower morbidity. Holmium laser enucleation of the prostate (HoLEP) was the first endoscopic enucleative technique described.8 This technique has been compared with OP and TURP in various randomised controlled trials, yielding at least comparable outcomes and a favourable safety profile.9 The use of expensive high-energy holmium laser equipment and a steep learning curve, however, have limited the extensive application of HoLEP worldwide. There has also been a significant risk of bladder injury associated with the use of the mechanical tissue morcellator that is required for HoLEP.
The use of normal saline as an irrigant was made possible by the introduction of bipolar devices. As a result, the risk of TURP syndrome has been virtually eliminated, and bipolar TURP has been widely adopted for resection of larger prostates with longer operating times. The use of a bipolar device in endoscopic enucleation was first reported by Neill et al,10 and bipolar TUERP requires no additional devices in comparison with bipolar TURP. Moreover, the sheath of the resectoscope is used for mechanical enucleation of the adenoma along the plane of the surgical capsule, instead of the holmium laser used in HoLEP. The subtotally enucleated adenoma is then resected into chips by the loop electrode, and the use of a mechanical tissue morcellator is eliminated.
The nomenclature for this procedure has not been standardised, with terms such as TUERP, plasmakinetic enucleation of the prostate, and bipolar plasma enucleation of the prostate reported in the literature. All of these names generally refer to the same procedure with minor differences. The term ‘bipolar TUERP’ is used in this article.
Several modifications in technique and equipment since the introduction of bipolar TUERP have been suggested. For example, a spatula-like enucleation loop, combined with a loop electrode for haemostasis, was introduced by Olympus and is especially designed for this procedure. Alternatively, the use of thick loop electrodes and button electrodes has been described in some series to facilitate the enucleation process. Based on personal experience with these different loops, the alternative loops with different designs are generally stronger than the conventional loop electrode, and they can be used for mechanical enucleation without breakage. The use of the loop in performing enucleation, instead of the resectoscope sheath, also provides better, more direct visualisation during the enucleation process and potentially shortens the learning curve and improves the safety of the procedure, particularly in the early phase of learning. The resectoscope sheath, however, facilitates a shorter enucleation time without compromising safety with the surgeon’s experience. The initial technique adopted for bipolar TUERP was the ‘three-lobe’ technique. This procedure starts with deep incisions down to the surgical capsule at the 5 and 7 o’clock positions from the bladder neck to the verumontanum, with an additional incision at the 12 o’clock position also reported. The median and lateral lobes of the prostate are then subtotally enucleated and resected in sequence. Some authors have reported ‘hybrid’ techniques, with the median lobe resected as in conventional TURP and only the lateral lobes enucleated. It has also been noted that deep incisions might not be necessary, as the surgical capsule plane can generally be identified with a small incision immediately proximal to the verumontanum, and the whole gland can be enucleated without separation of the lobes. This procedure avoids the bleeding associated with deep incisions of the bladder neck and adenoma although a small lobe can sometimes be difficult to enucleate after separation of the lobes. The 12 o’clock incision has been mostly abandoned, as has also been advocated for HoLEP. The anterior commissure, particularly the distal part, has been preserved to decrease the rate of transient urinary incontinence postoperatively. The technique used in our centre is currently the most widely practised among different centres.
Several series have reported the perioperative outcomes of bipolar TUERP using similar surgical techniques. Only the largest series from each centre was included for comparison; most of the published series are from China. The results of our series were compared with the TUERP arms of the various published series; this list of series and a comparison of the preoperative parameters are listed in Table 2. After the first published article by Neill et al10 comparing HoLEP and bipolar TUERP in 2006, Liu et al11 published the largest series with 1600 patients in 2010. Zhao et al12 and Liao and Yu13 followed by comparing bipolar TUERP and TURP in medium-sized prostates. Kan et al14 compared bipolar TUERP and TURP in large prostates, and Rao et al,15 Ou et al,16 Geavlete et al,17 and Chen et al18 compared bipolar TUERP with OP. The operative and early postoperative outcomes of bipolar TUERP from various studies are listed in Table 3.
Table 2. Comparison of preoperative parameters of patients with transurethral enucleation and resection of the prostate
The operating time was generally longer when the preoperative TRUS volume and resected prostate weight increased. Although Liu et al11 reported enucleation and resection times without reporting the total operating time, the preoperative TRUS volume and resected prostate weight were comparable between Liu et al’s report11 and our series. In addition, the resection time was prostate-size–dependent, and a resection efficacy of approximately 1 g/min was reported in both Liu et al’s report11 and our series. The enucleation time was less size-dependent, varying from 10 to 30 minutes, despite the large range of prostate sizes in our series.
The decrease in haemoglobin of approximately 10 g/L was reported for both medium-sized and large prostates. Early control of denuded vessels during the enucleation process made the removal of large glands possible, with minimal blood loss during the resection process.
The catheterisation and hospitalisation times varied greatly among the series evaluated. Longer times for both have typically been reported in series from China.11 12 13 15 16 18 In addition, no standard protocols were stated in most of the series, and the decision for catheter removal and hospital discharge were at the discretion of the surgeons. We report short catheterisation and hospitalisation times with the adoption of the same protocol as TURP in our institution. Specifically, bladder irrigation was stopped on postoperative day 1, the catheter was removed, and the patient was discharged from the hospital on postoperative day 2. A total of 92.9% of the patients (26 of 28 patients) complied with the postoperative protocol.
Postoperative TRUS volume was rarely reported by the series despite the consistent reporting of preoperative volume. This lack of reporting reflects the difficulty in accurately estimating residual tissue volume by TRUS, as illustrated by the postoperative TRUS photo shown in Figure 2. In addition, the central cavity remaining after TUERP can lead to overestimation of the prostate volume with the application of the traditional ellipse formula. Instead, preoperative estimation of the peripheral zone volume, obtained by subtracting the volume of the central zone from the total prostate volume, may represent a better method for estimating the residual tissue volume after TUERP. A decrease in TRUS volume of approximately 70% after TUERP was consistently reported, despite the pitfalls of postoperative TRUS measurements.
It has also been shown from the experience of HoLEP that a reduction in PSA level correlated with the amount of prostate tissue removed.19 Thus, serum PSA may serve as a better surrogate marker in the estimation of postoperative residual tissue volume. A postoperative PSA level of approximately 1 ng/mL and a decrease in PSA by >70% were commonly reported in most of the series.
Adverse events were poorly and inconsistently reported, as shown in Table 4. The standard Clavien classification was not adopted. There was no Clavien grade 3 or 4 complication in our series. The transfusion rate was low, with the exception of the series by Ou et al,16 and clot retention was rare. The rate of urinary tract infection ranged from 2% to 7.3%, and the re-catheterisation rate was <5%. Major complications were not common but did occur, as reported by Kan et al14; four admissions to intensive care units and nine conversions to other procedures were reported in this series of 74 patients. The rate of urethral stricture or bladder neck stenosis was low and comparable with conventional TURP as reported by the series by Liu et al.11 Long-term outcome was not reported in our series due to the short duration of follow-up. Temporary urinary incontinence was the major concern with enucleative procedures, and OP resulted in temporary urinary incontinence in approximately 10% of cases. The reporting of transient urinary incontinence after TUERP was poor and did not feature in three of the nine series analysed. Furthermore, the definition, timing, and severity of urinary incontinence were not stated in the other studies. Dramatic changes in the symptomatology of the patients over time following benign prostatic hyperplasia–related surgery, however, likely explain the difficulty in defining transient urinary incontinence. In our experience, transient urinary incontinence is not uncommon after TURP, although it is difficult to differentiate the type of urinary incontinence, stress, urge or mixed, by history or urodynamic studies. The natural history of this phenomenon has rarely been reported in the literature. It was interesting to note that the rate of transient urinary incontinence was much higher for the TURP group (16.1%) compared with the TUERP group (7.5%) in the series by Liao and Yu.13 In our experience, 17.9% of patients reported episode(s) of urinary incontinence at any time point after TUERP; the rate of transient urinary incontinence was 10.7% and 3.6% at 1 and 3 months postoperatively, respectively. Patients who had transient urinary incontinence used two pads daily on average, and all cases of transient urinary incontinence subsided by 4 months. Further investigations with, for example, measurement of pad weight and urodynamic studies will better delineate the cause and natural history of postoperative transient urinary incontinence. There is currently no predictive factor identified for the phenomenon.
Table 4. Comparison of preoperative parameters of patients with transurethral enucleation and resection of the prostate
Comparison of outcome and complications between patients with and without urinary retention was limited by the small patient number in our study. No significant difference between outcome and complications was identified even though patients with retention were significantly older.
A learning curve of 50 cases was reported for HoLEP,20 and this learning curve was expected to be shorter for bipolar TUERP. The instrumentation for TUERP should be familiar to an endourologist experienced in TURP because no additional devices are required. Xiong et al21 analysed the learning curve of bipolar TUERP. The ratio of conversion to conventional TURP decreased after 30 cases, and the efficiency of enucleation and resection increased with accumulative experience after 50 cases. Our series showed that the early postoperative outcomes were comparable to those of large series after approximately 35 cases, without an increase in adverse events. The findings were based on analysis of the learning curve of a single surgeon and may not be applicable to all surgeons. Nevertheless, an estimation of a learning curve in a magnitude of 30 to 50 cases seems reasonable and serves as a valuable reference.
Our study suggests that bipolar TUERP is a safe technique for prostates of any size. This procedure should become the endourological equivalent to open adenomectomy, with fewer complications and shorter convalescence. This technique can also be acquired safely with a relatively short learning curve.
1. AUA Practice Guidelines Committee. AUA guideline on management of benign prostatic hyperplasia (2003). Chapter 1: Diagnosis and treatment recommendations. J Urol 2003;170:530-47. CrossRef
2. Madersbacher S, Lackner J, Brössner C, et al. Reoperation, myocardial infarction and mortality after transurethral and open prostatectomy: a nation-wide, long-term analysis of 23,123 cases. Eur Urol 2005;47:499-504. CrossRef
3. Mamoulakis C, Ubbink DT, de la Rosette JJ. Bipolar versus monopolar transurethral resection of the prostate: a systematic review and meta-analysis of randomized controlled trials. Eur Urol 2009;56:798-809. CrossRef
4. Liu C, Zheng S, Li H, Xu K. Transurethral enucleative resection of prostate for treatment of BPH. Eur Urol 2006;5 Suppl:234. CrossRef
5. Rassweiler J, Teber D, Kuntz R, Hofmann R. Complications of transurethral resection of the prostate (TURP)—incidence, management, and prevention. Eur Urol 2006;50:969-79; discussion 980. CrossRef
6. Oelke M, Bachmann A, Descazeaud A, et al. EAU guidelines on the treatment and follow-up of non-neurogenic male lower urinary tract symptoms including benign prostatic obstruction. Eur Urol 2013;64:118-40. CrossRef
7. Freyer PJ. Total enucleation of the prostate. A further series of 550 cases of the operation. Br Med J 1919;1:121-120.2.
8. Gilling PJ, Kennett KM, Fraundorfer MR. Holmium laser enucleation of the prostate for glands larger than 100 g: an endourologic alternative to open prostatectomy. J Endourol 2000;14:529-31. CrossRef
9. Ahyai SA, Gilling P, Kaplan SA, et al. Meta-analysis of functional outcomes and complications following transurethral procedures for lower urinary tract symptoms resulting from benign prostatic enlargement. Eur Urol 2010;58:384-97. CrossRef
10. Neill MG, Gilling PJ, Kennett KM, et al. Randomized trial comparing holmium laser enucleation of prostate with plasmakinetic enucleation of prostate for treatment of benign prostatic hyperplasia. Urology 2006;68:1020-4. CrossRef
11. Liu C, Zheng S, Li H, Xu K. Transurethral enucleation and resection of prostate in patients with benign prostatic hyperplasia by plasma kinetics. J Urol 2010;184:2440-5. CrossRef
12. Zhao Z, Zeng G, Zhong W, Mai Z, Zeng S, Tao X. A prospective, randomised trial comparing plasmakinetic enucleation to standard transurethral resection of the prostate for symptomatic benign prostatic hyperplasia: three-year follow-up results. Eur Urol 2010;58:752-8. CrossRef
13. Liao N, Yu J. A study comparing plasmakinetic enucleation with bipolar plasmakinetic resection of the prostate for benign prostatic hyperplasia. J Endourol 2012;26:884-8. CrossRef
14. Kan CF, Tsu HL, Chiu Y, To HC, Sze B, Chan SW. A prospective study comparing bipolar endoscopic enucleation of prostate with bipolar transurethral resection in saline for management of symptomatic benign prostate enlargement larger than 70 g in a matched cohort. Int Urol Nephrol 2014;46:511-7. CrossRef
15. Rao JM, Yang JR, Ren YX, He J, Ding P, Yang JH. Plasmakinetic enucleation of the prostate versus transvesical open prostatectomy for benign prostatic hyperplasia >80 mL: 12-month follow-up results of a randomized clinical trial. Urology 2013;82:176-81. CrossRef
16. Ou R, Deng X, Yang W, Wei X, Chen H, Xie K. Transurethral enucleation and resection of the prostate vs transvesical prostatectomy for prostate volumes >80 mL: a prospective randomized study. BJU Int 2013;112:239-45. CrossRef
17. Geavlete B, Stanescu F, Iacoboaie C, Geavlete P. Bipolar plasma enucleation of the prostate vs open prostatectomy in large benign prostatic hyperplasia cases—a medium term, prospective, randomized comparison. BJU Int 2013;111:793-803. CrossRef
18. Chen S, Zhu L, Cai J, et al. Plasmakinetic enucleation of the prostate compared with open prostatectomy for prostates larger than 100 grams: a randomized noninferiority controlled trial with long-term results at 6 years. Eur Urol 2014;66:284-91. CrossRef
19. Tinmouth WW, Habib E, Kim SC, et al. Change in serum prostate specific antigen concentration after holmium laser enucleation of the prostate: a marker for completeness of adenoma resection? J Endourol 2005;19:550-4. CrossRef
20. Shah HN, Mahajan AP, Sodha HS, Hegde S, Mohile PD, Bansal MB. Prospective evaluation of the learning curve for holmium laser enucleation of the prostate. J Urol 2007;177:1468-74. CrossRef
21. Xiong W, Sun M, Ran Q, Chen F, Du Y, Dou K. Learning curve for bipolar transurethral enucleation and resection of the prostate in saline for symptomatic benign prostatic hyperplasia: experience in the first 100 consecutive patients. Urol Int 2013;90:68-74. CrossRef