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

In Hong Kong, as in most developed countries, one of the most common types of burn injury that requires hospital admission is scald. If we view a scald as a ‘contact burns due to a liquid’, the severity of burn injury will be a function of starting temperature, contact time, and the thermal capacity of the causative agent. In a previous study performed in our centre, it was hypothesised that both viscosity and thermal capacity of the agent were important factors in prolonging heat exposure of the skin.1 Patients who were scalded by congee (a popular local dish where rice with excess water is simmered until a porridge-like mixture is formed) were more likely to require surgery than those scalded by hot water (31% vs 14%, respectively).1 This may be partly explained by congee’s greater thermal capacity (causing more effective heat transfer) and viscosity (prolonging contact time).

In order to minimise the severity of burn injury, it is essential that effective first-aid measures be properly taught to the public. The two basic principles are to stop the burning process and to cool the burn wound. As patients are usually wearing clothes and underwear at the time of scalding, the most effective way to achieve the first objective is to remove the involved clothing as quickly as possible. For many reasons, however, this straightforward act is sometimes not performed in a timely manner (eg scald occurring in a restaurant where a patron is too embarrassed to take off his/her clothing or simple ignorance of first aid). It is in this context that a simple model was designed in an attempt to demonstrate the importance of prompt clothing removal after scalding has occurred.

In the first part of our study, the effect of delayed clothing removal on skin temperature was assessed using our experimental burn model. As in our centre’s previous study, a piece of Allevyn foam dressing (Smith & Nephew Medical Limited, Hull, England) served as our skin model. It was placed on a metal plate above a heated water bath (JP Selecta, Barcelona, Spain) until its surface temperature reached 34°C when measured using the Raytek non-contact thermometer (Raytek Corporation, Santa Cruz [CA], US) 1 cm above the centre, in an attempt to simulate ‘skin temperature’. Cotton underwear was then placed on top of the Allevyn dressing. The scalding agents (hot water and congee) were boiled to 88°C and poured onto the model. The underwear was then removed at various times following the ‘injury’ (10, 20, 30, 60, and 120 seconds). The temperature of the Allevyn was immediately measured at the aforementioned position by the same observer starting from the moment of clothing removal and every 10 seconds thereafter for 2 minutes. The same procedure was repeated 3 times for each time interval, and the mean value was taken as the final data point. Standard cooling curves were subsequently plotted using these average data points along with calculated standard deviations.

In the second part, all patients admitted to our unit via the emergency department with scalding by congee over a 10-year period (January 2005 to December 2014) were first retrospectively identified from our hospital’s computer records. Clinical notes were physically traced in order to examine the admission details as well as subsequent management of the patients.

The results of our experimental burn model are shown in Figures 1 and 2. The temperature of the Allevyn partly reflected the amount of heat transferred from the causative agent. A higher cooling curve translated to a higher skin temperature (hence higher potential injury). Examination of the cooling curves revealed a difference in the behaviour of hot water and congee. With hot water, the 10-second cooling curve was lower than its counterparts (the 20-, 30-, 60-, and 120-second cooling curves) that essentially lie along the same curve above. With congee, the cooling curves appeared to lie progressively higher with further delay in clothing removal—the 30- and 60-second cooling curves were in a significantly higher position compared with the 10- and 20-second cooling curves, while the 120-second cooling curve was higher than that of 30 and 60 seconds.

During the study period, 35 patients were admitted (21 males, 14 females) for scalding by congee. Of note, two paediatric patients were admitted to the Intensive Care Unit (ICU) for initial resuscitation due to the extent of their burns (16% and 20% of total body surface area [TBSA]) and one patient was discharged against medical advice after being admitted for 1 day. The demographic distribution, percentage of TBSA burnt, and length of hospital stay are shown in Table 1. On average, our unit admitted two to four cases per year except in 2010 when there were seven cases. Although most of these congee scalds were relatively minor as evidenced by the small mean percentage of TBSA involved and a mean hospital stay of approximately 11 days, more than two thirds of patients were children. Table 2 shows the statistics for clothing removal/first-aid measures and the subsequent management. While most of our patients received some form of first aid, it was found that the act of clothing removal was not documented in the majority (27/35) of cases. Despite an extended review period, it is recognised that the sample size remains small. Consequently, only descriptive statistics could be shown and a formal statistical analysis could not be conducted for this retrospective review. Nonetheless, those patients who received first aid or in whom clothing was removed (or both) did seem to fare better than those without.

This was a preliminary study of the potential relationship between the time of clothing removal and depth of scald burns. While ‘cooling the burn wound’ is an important step in the first-aid process, it must be recognised that ‘stop the burning process’ should take precedence in order to maximise the benefit of first aid. To this end, the contact time of the offending agent with the skin should be minimised. A cooling study was performed by our centre in 2006 to examine the cooling curves of different food/drinks. Of seven common agents examined for their rate of cooling, congee cooled significantly slower compared with the other agents (eg tea, coffee, noodles, etc).1 It was also shown that a higher percentage of patients required surgery if scalded by congee compared with hot water. As congee is a common dish for children in Hong Kong from which scalds could lead to greater morbidity, congee was specifically selected for further investigation.

We showed that delay in clothing removal could increase the severity of scald burns by congee as demonstrated by the increased temperature in our skin model when clothing removal was delayed. The experiment did not persist beyond a delay of 120 seconds since clothing would generally be removed within that time or not at all.

Hot water was first examined as a ‘control’ compared with congee. In our results, the cooling curves of 20, 30, 60, and 120 seconds essentially lay along the same curvature while the 10-second cooling curve was significantly lower. This implies that if clothing removal occurs within the first 10 seconds, the surface temperature of our skin model would be significantly lower at all subsequent time points; thus in hot water scalds, immediate clothing removal may prove to be the most beneficial.

On subsequent examination of the effect of congee, the cooling curves behaved differently—there appeared to be a stepwise progression from the 10- and 20-second curves to the 30- and 60-second curves, and then finally to the 120-second curve. Observation of the error bars showed that the 10- and 20-second curves did not differ significantly and this also applied to the 30- and 60-second curves. The 120-second curve generally lay significantly above the rest; therefore, it appears that delay in clothing removal significantly affected the surface temperature of our skin model—removing it within the first 20 seconds may lead to a less-severe injury compared with a 30- to 60-second delay, which in turn is better than a 2-minute delay.

At the time points between 120 and 140 seconds where all the curves have overlapping temperature measurements, one is able to observe that there are three distinct ‘tiers’ (10/20, 30/60, 120 seconds). In our model, heat was transferred to the Allevyn with the garment initially acting as a ‘barrier’ since the underwear was removed before the congee could soak through, resulting in a lower temperature detected. By increasing the time interval, the garment increasingly acted as a ‘reservoir’ for heat as the congee gradually soaked through. Due to the greater thermal capacity and viscosity of congee compared with water, the temperature of the entire ‘congee-underwear-Allevyn complex’ was maintained with more heat being transferred to the Allevyn for the same given period, resulting in a higher temperature detected. Overall, our findings corroborate our hypothesis that with a viscous agent such as congee, the severity of scald burns could potentially be reduced with earlier removal of clothing as evidenced by lower temperatures detected in our skin model.

Although our experimental study did succeed in demonstrating effects in our skin model, the ideal model for this study would be human skin (eg cadaveric or surgically excised) but it is rather difficult to source in practice. For the sake of scientific reproducibility, Allevyn was used instead (a bilayer dressing material with an outer waterproof layer analogous to the epidermis along with an inner absorbent sponge layer). Although Allevyn does not exactly mimic the ‘in-vivo’ behaviour of human skin per se, this model allows the opportunity for comparative study. Another potential improvement of our model is to set up a thermometer to measure the skin model’s temperature underneath the surface so that temperature can be tracked while clothing is still in place. In this experimental model, we were only able to measure the temperature after the garment was removed. Having continual temperature monitoring would enable us to plot ‘complete’ curves starting from 0 second onwards for all five time intervals, and the temperature change both before and after clothing removal could be more accurately depicted. In an early study by Moritz and Henriques,2 porcine skin was found to bear a remarkable resemblance to human skin. If the experiment can be repeated with porcine skin along with improved accuracy of temperature measurement (ie setting up a thermometer intradermally within the porcine skin), the validity of our conclusions may be strengthened. Nonetheless, this experimental study lends support to the notion that timely clothing removal before first-aid application may reduce the severity of burn injury.

The paediatric population (especially those under the age of 2 years) is particularly susceptible to scalds with a male preponderance.3 4 5 This age-group is particularly vulnerable as it is an age of great curiosity about the environment (hence the tendency to grab/tip over things) but limited motor development does not allow a child to move away from danger, such as a falling bowl of congee. It is also compounded by the fact that children have relatively thinner skin and this results in more significant injuries. Our statistics from this congee scald review do not deviate much from our centre’s previous experience: slightly over 50% of our patients were aged 2 years or younger with a male predominance, and 24 out of 35 patients were within the paediatric age-group.

It seems almost intuitive that clothing should be removed as soon as possible whenever a scald burn is sustained. This may not necessarily be so. In a recent UK study where parents were interviewed and asked about first-aid measures they would provide for a child with a large scald, 61% of parents failed to state that clothes should be stripped and several thought that it would cause further skin damage.6 The question of whether removing clothing would cause further skin damage is commonly asked by parents of paediatric burn patients admitted to our unit.

In our retrospective review, analysis of the efficacy of clothing removal was hindered because such action was not recorded in the majority of cases (27 out of 35). The location of burns was further studied: in 10 of these 27 cases, injury occurred over an exposed area (eg face and hands), while the remaining 17 cases could have benefited from clothing removal. In six of eight patients where clothing was removed, the exact timing was not documented. Such lack of documentation demonstrates the benefit of education about prehospital treatment for both the public and frontline medical staff. An increased awareness of correct prehospital treatment of burns would mean relevant questions are asked during history taking, and this would facilitate proper documentation and subsequent patient management. As mentioned in the Results section, our sample size did not permit any meaningful statistical analysis to be carried out pertaining to the potential usefulness of early clothing removal in reducing morbidity from scalds. Nonetheless, it appeared that those patients who had clothing removed fared quite well. One exception was a 50-year-old man with a relatively larger scald (TBSA, 13%) involving the face/neck/chest/bilateral upper limbs who eventually required skin grafting.

After the burning process is stopped, the next logical step is to cool the burn by applying first aid. Ideally, first aid for burns should provide pain relief and reduce potential morbidity associated with the injury. Although many first-aid treatments to cool burns have been studied, the application of cold water has the strongest supporting evidence. Currently, an ‘adequate’ first aid is defined as 20 minutes of running tap water over the burn within the first 3 hours of injury.7 Apart from removing heat energy from the damaged tissue, the benefits of cooling continue and include decreased oedema formation, preservation of dermal perfusion, decreased inflammatory response, and improved wound healing.8 9 Regrettably, our centre’s previously published paper showed that first aid was applied only to half of our paediatric patients.5 In our current review, although 27 patients received some form of first aid, only 15 (43%) patients received cool-water treatment of variable duration. The duration of first aid with cool water was either not recorded or fell short of the recommended 20 minutes. This state of affairs is certainly unsatisfactory and more public education is warranted.

Our data once again do not enable formal statistical analysis, and a number of factors such as percentage of TBSA burnt and depth of burn would affect the eventual outcome of our patients. Nonetheless, it does appear that those who received first aid required fewer surgeries. Detailed analysis of those who received no first aid revealed that none of the burns was classified as major even though four out of eight patients underwent skin grafting. In patients who received cool water as first aid, only one needed surgery—a 14-month-old boy with a TBSA of 13.5% burnt involving the left flank and bilateral lower limbs. In patients who received other types of first aid, three required surgeries (one of whom had a major burn and required ICU admission). Of note, for patients in whom clothing was removed and first-aid measures applied, all were managed conservatively. Although definite conclusions from our data cannot be drawn, clothing removal and first aid do seem to have beneficial effects, especially in smaller burns that constitute most of our case load. It is our hope that with better medical documentation and education of our frontline staff, the quality of our future data can be enhanced with a view to facilitate formal data analysis.

Before a simple and effective message can be delivered to the public about post-burn prehospital management, it is equally important to consider the local food characteristics. For instance, the population of Hong Kong may find it easier to relate burns to congee or cup noodle than burns to coffee. Since half of our paediatric burn patients received no first aid upon admission, simply emphasising the importance of ‘stop the burning process’ may reduce potential morbidity. Our experimental skin model showed that earlier clothing removal post-burn reduces skin temperature and thus beneficial. Although our retrospective review was unable to reflect any concrete statistical results, it was certainly suggestive of the potential usefulness of proper prehospital management. Last but not least, the financial cost of managing acute uncomplicated minor paediatric scalds is significant (including hospital beds, theatre visits, dressings, medications etc). This is an important social and economic issue since burns sustained by children often require many years of follow-up for scar management and psychosocial support.10 If preventive measures fail and accidents occur, it is in the best interests of the public to understand how to minimise morbidity.

1. Chiu TW, Ng DC, Burd A. Properties of matter matter in assessment of scald injuries. Burns 2007;33:185-8. Crossref

2. Moritz AR, Henriques FC. Studies of thermal injury: II. The relative importance of time and surface temperature in the causation of cutaneous burns. Am J Pathol 1947;23:695-720.

3. Ray JG. Burns in young children: a study of the mechanism of burns in children aged 5 years and under in the Hamilton, Ontario Burn Unit. Burns 1995;21:463-6. Crossref

4. Dewar DJ, Magson CL, Fraser JF, Crighton L, Kimble RM. Hot beverage scalds in Australian children. J Burn Care Rehabil 2004;25:224-7. Crossref

5. Tse T, Poon CH, Tse KH, Tsui TK, Ayyappan T, Burd A. Paediatric burn prevention: an epidemiological approach. Burns 2006;32:229-34. Crossref

6. Graham HE, Bache SE, Muthayya P, Baker J, Ralston DR. Are parents in the UK equipped to provide adequate burns first aid? Burns 2012;38:438-43. Crossref

7. First aid. Australian and New Zealand Burn Association. Available from: http://anzba.org.au/care/first-aid/. Accessed Feb 2016.

8. Cuttle L, Pearn J, McMillan JR, Kimble RM. A review of first aid treatments for burn injuries. Burns 2009;35:768-75. Crossref

9. Wright EH, Harris AL, Furniss D. Cooling of burns: mechanisms and models. Burns 2015;41:882-9. Crossref

10. Griffiths HR, Thornton KL, Clements CM, Burge TS, Kay AR, Young AE. The cost of a hot drink scald. Burns 2006;32:372-4. Crossref

In Hong Kong, career opportunities in obstetrics and gynaecology (O&G) for medical graduates have undergone a roller-coaster ride over the last two decades. For O&G, the number of new trainees per year in the territory dropped from 11 in 2000 to a trough of five in 2004, followed by a steep rise to 22 in 2006.1 This trend was driven by multiple factors, notably the local birth rate and the influx of non-eligible person deliveries during this period.

According to the discussion paper ‘Manpower requirement and strategies for doctors’ published by the Hospital Authority in 2010, the annual intake requirement of O&G doctors was projected to be 20 for the period 2010 to 2016.2 Recruitment of O&G resident trainees in the latest Annual Recruitment Exercise in 2015 was unexpectedly difficult, however. Among the 24 O&G resident trainee posts available in the Annual Recruitment Exercise for Resident Trainees 2015/16,3 less than 50% were filled by May 2015 and 11 posts remained vacant in July 2015,4 thus creating a manpower crisis in O&G.

The trend of declining interest of medical graduates in pursuing O&G as a career has likewise been observed in overseas studies5 6 with various factors explored, such as perception of O&G clerkship,5 6 7 lifestyle,5 7 gender,5 7 8 and medicolegal issues.5 7 Local data on this topic are lacking. A literature search revealed one local study on specialty choice of medical graduates was conducted 20 years ago9 and another focused on the specialty of family medicine.10 In view of the recent manpower crisis, there was an urgent need to perform a study on this topic as it might guide manpower planning and formulation of solutions.

The objectives of this study were to evaluate the career interest of the most recent medical graduates in Hong Kong, especially their level of career interest in O&G, and to identify key influential factors for career choice and career interest in O&G.

This was a cross-sectional questionnaire survey. All medical graduates of the University of Hong Kong (HKU) and the Chinese University of Hong Kong (CUHK), who graduated in 2015 and attended the pre-internship lectures held in their medical school on 5 and 8 June 2015 respectively, were invited to participate. This study was approved by the Institutional Review Board of the University of Hong Kong / Hospital Authority Hong Kong West Cluster and the Joint Chinese University of Hong Kong–New Territories East Cluster Clinical Research Ethics Committee.

All medical graduates who attended the lectures were given an information sheet and consent form. Those who consented were invited to complete a self-administered anonymous questionnaire in English that was collected on the same day by the investigator.

The questionnaire comprised 14 questions in three sections: (1) demographic data, (2) current view on career choice, and (3) view on O&G as a career. The second section explored the first three choices of specialty as a career and the extent of influence of 28 items on career choice. The third section evaluated the current level of career interest in O&G, the extent of influence of 28 items on career interest in O&G, and the impact on the level of career interest in O&G of five possible changes. The 28 items were potential influential factors that were identified from overseas studies5 6 7 8 in the following categories: (1) ability, interest, and personality, (2) clerkship experience, (3) influence of family, peers, and seniors, (4) medicolegal issues, (5) nature of the specialty, (6) other experience related to the specialty, (7) career prospects, (8) religion, and (9) training and working conditions.

All statistical analyses were performed using PASW Statistics 18, Release Version 18.0.0 (SPSS Inc., 2009, Chicago [IL], US). Categorical data were presented as counts and percentages. Continuous data were expressed as median (interquartile range) as they were highly skewed. Chi squared test and Fisher’s exact test were used to compare categorical data. For continuous data, as they were highly skewed, a non-parametric test (Mann-Whitney U test or Kruskal-Wallis H test) was used. Results with a P value of <0.05 were considered statistically significant.

In order to investigate the structure of the potential factors influencing career choice and career interest in O&G, an exploratory factor analysis by the extraction method of principal component analysis with varimax rotation was performed. An exploratory factor analysis was performed instead of a confirmatory factor analysis because the factor structure was uncertain. The number of factors to be extracted was based on the result from the scree-plots and the Kaiser’s eigenvalue criterion (eigenvalue >1). Missing values were excluded listwise. The quality of the factor analysis models was assessed by Bartlett’s test of sphericity and the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy.

In June 2015, there were 323 new medical graduates in Hong Kong, of whom 163 were from HKU and 160 were from CUHK. Among them, 283 attended the pre-internship lectures (157 HKU graduates and 126 CUHK graduates); 238 participants were recruited who constituted 73.7% of all 323 medical graduates and 84.1% of 283 medical graduates in the lectures. Five questionnaires were excluded from analysis due to significant missing data (four missing the level of career interest in O&G and one missing the extent of influence of factors for career interest in O&G). Thus, 233 questionnaires were analysed.

Table 1 shows the demographic data of participants. There were 229 (98.3%) participants who planned to undergo specialty training after internship. Table 2 shows the distribution of their first three choices of specialty. There were 13 (5.7%) participants who indicated O&G as their first choice of specialty and 37 (16.2%) participants who listed O&G among their first three choices of specialty, of whom 29 (78.4%) were female and eight (21.6%) were male.

On a scale of 1 (no interest at all) to 10 (extremely interested), the median (interquartile range) score for the current level of career interest in O&G was 3 (1-6). The association between demographics and level of career interest in O&G is shown in Table 3. Gender was significantly associated with level of career interest in O&G with females more interested than males. There was also a statistically significant association between gender and listing O&G among the first three choices of specialty (P=0.001).

In the exploratory factor analysis, the values of KMO measure of sampling adequacy were >0.6. The P values of Bartlett’s test of sphericity were both <0.05. These figures indicated that the use of factor analysis was appropriate to investigate the underlying factors in this study. Eight underlying factors were extracted for career choice and seven underlying factors were extracted for career interest in O&G. All had an eigenvalue of >1, implying that these factors were meaningful. The total percentage of variance explained by the eight underlying factors for career choice and that for the seven underlying factors for career interest in O&G were both 68.0%, meaning these factors could explain 68.0% of the condition.

The three factors with the highest percentage of variance explained were identified as the key influential factors. The key influential factors for career choice and career interest in O&G are shown in Table 4.

Comparison of gender with the median of key influential factors for career choice and career interest in O&G revealed no difference except for clerkship experience. This was the strongest key influential factor for career interest in O&G. The median rating by female participants was 21 (18-24) and that for male participants was 19 (16-22) [P=0.021].

Impact on career interest in O&G of five potential changes was evaluated (Fig). Decreasing the frequency of on-call duty and better remuneration had the highest positive impact. The option of part-time training with a longer training period had the lowest positive impact but highest negative impact.

To our knowledge, this is the first study in Hong Kong to investigate the career interest of medical graduates in O&G and the influential factors. It was satisfactory to have a participation rate of 73.7% of the 323 medical graduates in Hong Kong this year.

Medicine and surgery were the most popular specialty choices. This is consistent with previous local studies.9 10 On the other hand, O&G ranked eighth when combined with the first three choices and sixth as the first choice. This is a reduction from previous local studies in which its popularity was ranked as third9 and fourth10, respectively.

The level of career interest in O&G was low with a median score of 3 (out of 10). Female participants had a higher career interest in O&G than their male counterparts, which is consistent with overseas studies.8

The key influential factors for career choice and career interest in O&G were consistent, namely clerkship experience, working style, and career prospects.

Working style was the strongest key influential factor for career choice, with the highest percentage of variance explained (26.4%). The items included in this factor had a similar coefficient of around 0.8, reflecting that each item made a similar contribution to this factor. This factor was important to both female and male medical graduates with no statistically significant gender difference. The focus on a work-life balance has attracted increasing attention, not just locally but globally, and is not unique to the medical profession. There were 149 (63.9%) participants who considered a decreased on-call frequency would have a positive impact on their level of interest in O&G. This had the highest positive impact among the five possible changes explored. Efforts should be made to avoid the vicious cycle of lack of manpower and increased on-call frequency in order to maintain the attractiveness of O&G to future generations of the profession. In addition, providing better remuneration, such as a special honorarium for extra on-call duties, might be explored, especially when manpower is insufficient. This was thought to increase career interest in O&G by 141 (60.5%) participants.

Clerkship experience was the strongest key influential factor for career interest in O&G with the highest percentage of variance explained (28.9%). Included in this factor, five items had a higher contribution with coefficients of above 0.6, namely (1) overall learning experience in O&G clerkship, (2) interaction with O&G interns and trainees, (3) interaction with O&G specialists, consultants, and professors, (4) hands-on experience in O&G clerkship, and (5) interaction with midwives and O&G nurses. This result is consistent with overseas findings of the important influence of perception of clerkship, including interaction with O&G staff, on career interest in O&G.5 6 It emerges that O&G nurses, midwives, interns, trainees, and specialists play a vital role in ensuring a positive O&G clerkship experience.

Noticeably, this was the only key influential factor with a significant gender difference (P=0.021): the clerkship experience of female participants had a more positive impact on their career interest in O&G compared with males. The reason behind this phenomenon was beyond the scope of this study. Interestingly, Chang et al6 suggested that male medical students were more likely to increase their level of career interest in O&G during the clerkship. Assuming this finding is applicable to our local students, further study of gender difference in clerkship experience may help improve the imbalanced gender ratio of O&G trainees. In our study, 126 (54.1%) participants agreed that a more balanced gender ratio of staff would have a positive impact on career interest in O&G.

It might be surprising to note that some items that might have been expected to be influential—such as interest in the specialty,10 compatibility with personality, and availability of training posts—were not among the key influential factors in this study.

There are several limitations of this study. First, recruitment of participants was affected by the attendance of medical graduates at the pre-internship lectures: 87.6% of all medical graduates. In addition, participation in this study was voluntary: 45 (15.9%) of 283 medical graduates who attended the lectures did not participate. The reason for non-participation was not documented. There may be potential response bias as those with a low career interest in O&G may have been less keen to participate. Other possible reasons for non-participation were unwillingness to disclose career interest or uncertain career interest. A similar study may be conducted with interns towards the end of the internship when career interest is usually clearer. There may be possible changes in career interest and influential factors at another stage of career development. Some external factors, such as the availability of training posts at the time of application, may become more influential at this juncture. Second, the findings have weak generalisability as there are many external factors that change with time and may be different in different parts of the world. It may be useful to conduct similar study periodically to obtain an updated trend to facilitate manpower planning. Third, the items of possible influential factors included in the questionnaire were not exhaustive and were selected with reference to previous studies.

The study confirmed the observation of a low level of career interest in O&G among medical graduates and the decreasing popularity of O&G as a career choice. Key influential factors for career interest in O&G and career choice included working style, clerkship experience, and career prospects. Encouragingly, there are modifiable elements among these factors, where improvement can be made by faculties, college, frontline staff, and hospital administrators.

We would like to thank Prof TP Lam and Dr Gilberto KK Leung, Institute of Medical and Health Sciences Education, Li Ka Shing Faculty of Medicine, The University of Hong Kong and Prof Tony KH Chung, Faculty of Medicine, The Chinese University of Hong Kong for their support and endorsement for this study to be conducted in the Faculties. We are grateful to Dr WC Leung, Chairman of Education Committee, The Hone Kong College of Obstetricians and Gynaecologists and the Chief of Service of our respective departments: Dr Anita Yeung, Prof Ernest HY Ng, and Dr TH Cheung, for their support. We would also like to acknowledge Dr Carmen KM Choi and Dr Daniel Wong for their helpful comments on the questionnaire.

1. Au Yeung SK. Impact of non-eligible person deliveries in obstetric service in Hong Kong. Hong Kong J Gynaecol Obstet Midwifery 2006;6:41-4.

2. Manpower requirement and strategies for doctors, Hospital Authority Administrative and Operational Meeting Paper No 693. Available from: http://www.ha.org. hk/haho/ho/cad_bnc/AOM-P693.pdf. Accessed 5 May 2015.

3. Posts available, Annual Recruitment Exercise for resident trainees 2015/16, Hospital Authority. Available from: http://www.ha.org.hk/ho/resident.htm. Accessed 5 May 2015.

4. Unfilled resident trainee post for Annual Recruitment Exercise 2015/16, Hospital Authority. Available from: http://www.ha.org.hk/ho/resident.htm. Accessed 3 Jul 2015.

5. Gariti DL, Zollinger TW, Look KY. Factors detracting students from applying for an obstetrics and gynecology residency. Am J Obstet Gynecol 2005;193:289-93. Crossref

6. Chang JC, Odrobina MR, McIntyre-Seltman K. Residents as role models: the effect of the obstetrics and gynecology clerkship on medical students’ career interest. J Grad Med Educ 2010;2:341-5. Crossref

7. Fogarty CA, Bonebrake RG, Fleming AD, Haynatzki G. Obstetrics and gynecology—to be or not to be? Factors influencing one’s decision. Am J Obstet Gynecol 2003;189:652-4. Crossref

8. Schnuth RL, Vasilenko P, Mavis B, Marshall J. What influences medical students to pursue careers in obstetrics and gynecology? Am J Obstet Gynecol 2003;189:639-43. Crossref

9. Lam TP. Specialty choice of medical graduates. Hong Kong Pract 1996;18:504-12.

10. Wu SM, Chu TK, Chan ML, Liang J, Chen JY, Wong SY. A study on what influence medical undergraduates in Hong Kong to choose family medicine as a career. Hong Kong Pract 2014;36:123-31.

Mushroom poisoning is a global phenomenon and can be a source of major mortality and morbidity. Although Hong Kong people consume a large volume of mushrooms, there are few clinical studies and reviews related to local mushroom poisoning.1 2 The diagnosis of mushroom poisoning should be based on clinical features, laboratory investigations, and mushroom identification. Due to the lack of leftover mushroom samples in most cases, emergency physicians and clinical toxicologists usually have to diagnose mushroom poisoning based on clinical syndromes alone without mushroom identification by mycologists. Diaz3 has reviewed and established the classification of mushroom poisoning based on the time of presentation and target organ systemic toxicity. With respect to the time of presentation, mushroom poisoning is categorised as early onset (<6 hours), late onset (6-24 hours), or delayed onset (>1 day). Early-onset toxicities include several neurotoxic, gastrointestinal, and allergic syndromes. Late-onset toxicities include hepatotoxic, accelerated nephrotoxic, and erythromelalgia syndromes. Delayed-onset toxicities include delayed nephrotoxic, delayed neurotoxic and rhabdomyolytic syndromes. Syndromic approaches guide earlier diagnosis and facilitate empirical treatment.3

In Hong Kong, scattered cases of mushroom poisoning are reported every year. In a report published by the Centre for Health Protection (CHP), there were 13 reported cases of wild mushroom poisoning between January 2002 and May 2005.1 Symptoms occurred 0.5 to 5 hours post-ingestion and included vomiting (100%), abdominal pain (100%), diarrhoea (69%), nausea (56%), dizziness (50%), sweating (37%), numbness (31%), palpitation (19%), malaise (13%), fever (13%), and headache (6%).2 Of these patients, seven required hospitalisation and all of them completely recovered. Individual cases of mushroom poisoning have been announced in CHP press releases from time to time. In another report, seven patients presented with mainly gastrointestinal symptoms after wild mushroom consumption between May 2007 and August 2010.2 There have been no reports of life-threatening wild mushroom poisoning in Hong Kong before this case series.

Since its establishment in 2005, the Hong Kong Poison Information Centre (HKPIC) has provided a 24-hour telephone consultation service to health care professionals in Hong Kong for poison information and clinical management advice. We are actively involved in the diagnosis and management of mushroom poisoning cases in local hospitals. Supported by Prof SW Chiu from the School of Life Sciences of The Chinese University of Hong Kong, mycological identification can be provided whenever mushroom samples are available in a poisoning case. The objectives of this study were to review the clinical features and mycological identifications in mushroom poisoning cases recorded by the HKPIC.

Mushroom poisoning cases recorded by HKPIC from 1 July 2005 to 30 June 2015 were retrospectively reviewed. Information on patient demographic details, clinical presentation, sources of mushroom, investigation results, mycological identification results, and clinical outcome were obtained. Descriptive statistics were used for data analysis.

During the 10-year study period, there were 67 cases of mushroom poisoning. All cases were reported from hospitals of the Hospital Authority. All patients were Chinese; 29 (43%) were male and 38 (57%) were female. The median age was 47 (range, 2-86) years. Of the 67 cases, 52 (78%) occurred between April and September when the climate in Hong Kong is optimal for mushroom growth. In 66 cases, the mushrooms were intentionally consumed as delicacies. There was one case of accidental ingestion, in which a 2-year-old boy ingested a wild mushroom in a park. Ingestion with recreational, suicidal, or malicious intent was not recorded in this case series.

The most common clinical presentation was gastrointestinal symptoms (Table 1). Symptoms mimicking gastroenteritis were the presenting feature in 60 (90%) patients. The diagnosis of these 60 patients included gastroenteritic mushroom poisoning in 38, cholinergic mushroom poisoning in five, food poisoning in eight, food allergy in two, and seven cases of amatoxin poisoning.

Neurological symptoms were also commonly reported: 20 (30%) patients presented with one or more symptoms including dizziness, numbness, hallucination, headache, or confusion (Table 1). Most of them (15 of 20 patients) presented with both neurological and gastrointestinal symptoms. Four patients presented with neurological symptoms including visual hallucinations, dizziness, generalised weakness, and malaise without gastrointestinal symptoms. They were subsequently diagnosed with hallucinogenic mushroom poisoning. One patient presented with malaise, muscle fasciculation, and profuse sweating that was diagnosed as cholinergic mushroom poisoning.

The symptom onset time was documented in 62 patients (Table 2). Among them, 53 patients developed symptoms within 6 hours of mushroom consumption (early-onset group). The median time of symptom onset was 2 hours post-ingestion (interquartile range [IQR], 2). Most patients (50 out of 53) presented with early-onset gastrointestinal symptoms. No severe clinical outcomes were observed in this group of patients and all recovered with symptomatic treatment and short duration of hospital care.

Symptoms developed 6 hours or more after mushroom consumption in nine patients (late-onset group). The median time of symptom onset was 11 hours post-ingestion (IQR, 2). This group of patients represented potentially life-threatening mushroom poisoning. All seven cases of amatoxin poisoning in this case series were found in this group. The remaining two cases included one case of hallucinogenic mushroom poisoning caused by Tylopilus nigerrimus, and one case of food poisoning.

The source of poisonous mushrooms was documented in 64 cases (Table 3). In 34 (51%) cases, the mushrooms were self-picked from a park, hillside, or roadside. The locations were usually close to the patient’s home. On the other hand, 14 (21%) cases purchased the mushrooms in Hong Kong and 16 (24%) purchased them in mainland China. All patients with amatoxin poisoning collected the mushroom from the wild. The source of mushrooms was not documented in three (4%) cases.

Mycological identification was achieved in 28 cases (Table 4). The diagnosis of amatoxin poisoning was confirmed by the presence of amatoxin and/or phallacidin in the urine of five patients.

The aim of this study was to describe the pattern of mushroom poisoning in Hong Kong. Four mushroom poisoning syndromes, together with food poisoning and food allergy, were identified to be the cause of all mushroom poisoning cases in this study. The typical clinical features and the management of the four local mushroom poisoning syndromes are summarised in Table 5.4

As an extensive urban city, commercially sold cultivated mushrooms are easily available and these are the mushrooms consumed by most Hong Kong citizens every day. Nonetheless the Chinese generally believes that wild-harvested products, including mushrooms, have higher nutritional and medicinal values. The risky behaviour of collecting and consuming wild mushrooms was considered to be rare in Hong Kong. This can be illustrated by the relatively few reported cases of poisoning during the study period. There are over 388 known species of mushroom in Hong Kong,5 of which fewer than 10% are edible, and a majority have unknown edibility. Although mushrooms are macroscopic organisms with visible morphological features, many mushroom species share a similar appearance and misidentification is common. There is no correlation between a particular morphological feature and poisonous nature of a mushroom species. Even with genus Amanita, there are edible and inedible species. There is no simple way to differentiate edible and poisonous mushroom species. Even in expert hands, mushroom identification frequently depends on the microscopic features that can usually be seen in a laboratory setting. Different edible and inedible or poisonous mushroom species can share a similar habitat and grow in close proximity in the wild. Collection of mixed species often happens. Cooking or other means of food processing cannot detoxify a poisonous mushroom. With the report of life-threatening amatoxin poisoning from ingestion of local wild mushrooms, Hong Kong citizens would be well advised to stop the risky behaviour of consuming self-picked mushrooms from the wild.

Consumption of poisonous mushrooms can cause various signs and symptoms, such as gastroenteritis, disturbances in central nervous system, and liver failure.3 As mushroom identification is usually not available early on in patient care, doctors should treat their patients according to the clinical syndrome (Table 54). An important predicting factor to consider is the latency from ingestion to onset of symptoms. The finding of our case series is compatible with overseas reports.3 6 Patients with early-onset symptoms, typically within 6 hours post-ingestion, all had a benign course of disease (Table 2). The mainstay of treatment is supportive, with intravenous fluids, antiemetic, antispasmodics, or analgesic for those patients who present with gastrointestinal symptoms.

Seven cases of amatoxin poisoning reported in this case series confirms the existence of deadly amatoxin-containing mushroom in our locality. Poisonous Amanita species have long been found in Hong Kong. Although local cases of amatoxin poisoning have not been reported in Hong Kong before 2013, it has been well-reported in mainland China.7 8 9 According to a report published by Guangzhou Municipal Centre for Disease Control and Prevention, there were 92 cases of mushroom poisoning with 13 deaths in the years 2002 to 2005.9 The reported species of mushroom involved in 70% of cases were the amatoxin-containing mushroom Amanita exitialis and the gastroenteritic mushroom Chlorophyllum molybdites.9 For amatoxin poisoning, the latency between ingestion and onset of symptoms was typically 6 to 24 hours.3 For our seven cases of amatoxin poisoning, this latency ranged from 8 to 12 (median, 11) hours. There were four male and three female patients, with a median age of 44 (range, 29-74) years. All cases presented with persistent vomiting and diarrhoea, deranged liver function tests, and were able to give a history of wild mushroom ingestion. Three were imported cases. Two of them ate wild mushrooms in South Africa and one patient ate wild mushrooms in China. In four patients, wild mushrooms were picked locally in the country park of the New Territories. One imported case presented to hospital in Hong Kong 5 days after wild mushroom consumption and died of multi-organ failure soon after hospital admission. The remaining six cases were managed according to overseas experience in the treatment of amatoxin poisoning.10 11 12 Treatment included intravenous silibinin, oral silymarin, intravenous N-acetylcysteine, oral multiple-dose activated charcoal, high-dose intravenous penicillin, and early charcoal haemoperfusion. Two local cases progressed to liver failure and required liver transplantation. The remaining four cases recovered with medical treatment. The diagnosis of amatoxin poisoning was confirmed by the presence of amatoxin and/or phallacidin in the urine in five patients. Mycological examination identified Amanita farinosa as the causative mushroom in a local incident with two patients (Table 4). This is the first report of this Amanita species in Hong Kong.

Hallucinogenic mushroom poisoning has not been previously reported locally. The clinical presentation of our four cases included dizziness, headache, generalised weakness and numbness. Three out of four patients presented with visual hallucination. Although one patient did not report any hallucinations, the patient was included as a suspected case of hallucinogenic mushroom poisoning based on compatible neurological symptoms following consumption of porcini. The symptom onset time was documented in two cases and was 2 hours and 10 hours post-ingestion. The source of mushroom was recorded in three cases as mainland China. In only one case was mycological identification performed (Table 4).

There were 13 cases of bolete poisoning in this case series. All cases were related to consumption of porcini. Porcini is considered to be a delicacy by many mushroom lovers. It includes a number of edible Boletus species, with Boletus edulis being the best known. Not all Boletus are edible, however, and mixing edible and inedible species is possible in wild mushroom harvesting. There are reports of bolete poisoning in English and Chinese literature.6 Bolete consumption has been associated with outbreaks of neuropsychiatric symptoms in Southwest China (eg Yunnan province) in recent years.13 14 15 According to these reports, consumption of inedible boletes typically presented with gastrointestinal and neurological symptoms including visual and auditory hallucination. In our case series, two out of 13 cases of bolete poisoning presented with neuropsychiatric symptoms without gastrointestinal symptoms. The two unrelated cases purchased mushrooms from Yunnan province. The first patient presented with numbness and weakness in all four limbs, dizziness, and malaise after mushroom consumption. The time of symptom onset was not documented and symptoms resolved on the same day as mushroom consumption. No mushroom sample was obtained for identification. The second patient developed dizziness, malaise, and visual hallucination 10 hours after mushroom consumption. Her symptoms resolved 48 hours post-ingestion. The causative mushroom was identified as T nigerrimus, an inedible bolete. Hallucinogenic mushroom poisoning caused by T nigerrimus has not been reported in the English literature.

In the cases of bolete poisoning, 11 out of 13 presented with gastrointestinal symptoms after mushroom consumption. Four patients purchased the mushrooms locally, and seven purchased the mushrooms in China. In most cases, the mushrooms were commercially packed as a product containing wild-harvested boletes in dried slices. Mycological identification was performed in 10 cases. These 10 cases represented six poisoning incidents. In five incidents involving eight patients, the wild-harvested porcini showed mixing up of edible porcini and inedible boletes. In the remaining one incident, the mushrooms were identified as an edible species, although microscopic examination revealed them to be rotten with dried worm and mold (Table 4). The diagnosis was food poisoning with possibly bacterial contamination of spoiled mushroom in this incident.

Most cases of mushroom poisoning in Hong Kong follow a benign course. Life-threatening cases of amatoxin poisoning are occasionally seen. Doctors should consider this diagnosis in patients who present with gastrointestinal symptoms whose onset is 6 hours or more after mushroom consumption. In this review, all patients with amatoxin poisoning picked the poisonous mushroom from the wild. Wild mushroom picking and consumption should be strongly discouraged.

1. Centre for Health Protection. Food poisoning associated with wild mushroom. Communicable Disease Watch 2005;2:41-2.

2. Chan TY, Chiu SW. Wild mushroom poisonings in Hong Kong. Southeast Asian J Trop Med Public Health 2011;42:468-9.

3. Diaz JH. Syndromic diagnosis and management of confirmed mushroom poisonings. Crit Care Med 2005;33:427-36. Crossref

4. Hoffman RS, Howland MA, Lewin NA, et al. Goldfrank’s toxicologic emergencies. 10th edition. New York: McGraw-Hill Education; 2015.

5. Chang ST, Mao XL. Hong Kong mushrooms. Hong Kong: The Chinese University Press; 1995.

6. Schenk-Jaeger KM, Rauber-Lüthy C, Bodmer M, Kupferschmidt H, Kullak-Ublick GA, Ceschi A. Mushroom poisoning: a study on circumstances of exposure and patterns of toxicity. Eur J Intern Med 2012;23:e85-91. Crossref

7. Jin LM, Li Q. 金連梅, 李群. [Analysis of food poisoning incidents during 2004 to 2007 in China] 2004-2007年全國食物中毒事件分析 [in Chinese]. [Disease Surveillance] 疾病監測 2009;24:459-61.

8. Guo SH, Liu FQ, Liu XY, Chen Y. 郭綬衡，劉富强，劉秀英，陳焱. [Analysis of food-borne disease incidents in Jiangsu, Zhejiang, Hunan, Hubei and Hebei Province during 2000 to 2002] 2000-2002年江蘇、浙江、湖南、湖北、河北五省食物源性疾病發病情况分析 [in Chinese]. [Practical Preventive Medicine] 實用預防醫學 2004;11:867-71.

9. Mao XW, Li YY, He JY, Jing QL. 毛新武，李迎月，何潔儀，景飲隆. [Investigation of mushroom poisoning in Guangzhou City from 2000 to 2005] 廣州市 2000-2005年蘑菇中毒調查 [in Chinese]. [China Tropical Medicine] 中國熱帶醫學 2007;7:166-7.

10. Ward J, Kapadia K, Brush E, Salhanick SD. Amatoxin poisoning: case reports and review of current therapies. J Emerg Med 2013;44:116-21. Crossref

11. Enjalbert F, Rapior S, Nouguier-Soulé J, Guillon S, Amouroux N, Cabot C. Treatment of amatoxin poisoning: 20-year retrospective analysis. J Toxicol Clin Toxicol 2002;40:715-57. Crossref

12. Giannini L, Vannacci A, Missanelli A, et al. Amatoxin poisoning: a 15-year retrospective analysis and follow-up evaluation of 105 patients. Clin Toxicol (Phila) 2007;45:539-42. Crossref

13. Ji X, Ma Z, Zhu H, Pan YZ. 及曉，馬征，朱輝，潘軼竹. [Report of two cases of mental disorder due to ingestion of Boletus speciosus] 小美牛肝菌所致精神障礙2例 [in Chinese]. [Clinical Journal of Psychiatry] 臨床精神醫學雜誌 2014;24:240.

14. Liu MW, Zhou H, Hao L, Zhang MQ. 劉明偉，周惠，郝麗，張明謙. [Analysis of 61 cases of boletus poisoning] 牛肝菌中毒61例分析 [in Chinese]. [Chinese Journal of Misdiagnosis] 中國誤診學雜誌 2008;8:111.

15. Zhou YJ, Wei GL, Chen GH. 周亞娟，魏桂蘭，陳桂華. [Investigation of Rhubarb boletus food poisoning] 一起黄粉牛肝菌食物中毒事件調查 [in Chinese]. [Occupational Health and Damage] 職業衛生與病傷2008;23:115-6.

Prostate cancer is the most common cancer diagnosed in men in developed countries. In the United States, there were 240 890 estimated new cases in 2011, accounting for 29% of all new cancers in men and over 33 000 deaths.1 According to the Hong Kong Cancer Registry in 2012, prostate cancer was the third most common cancer in men.2

The overall incidence of advanced-stage prostate cancer has declined in recent years, probably due to early detection and treatment following application of prostate-specific antigen (PSA) for prostate cancer screening.3 Nonetheless it has been shown that approximately 4% of patients present with metastatic disease at the time of diagnosis1 and 5% present with localised or regional disease that eventually metastasises.4

Bone is the major metastatic site of prostate cancer, and has been observed in 90% of patients during autopsy.4 Common sites of metastases include the vertebrae, pelvis, long bones, ribs, and skull. Bone metastases cause major morbidity in patients with prostate cancer. They weaken the structural integrity of bone, leading to an increased risk for skeletal-related events (SREs) such as pathological fracture, spinal cord compression, and severe bone pain requiring palliative radiotherapy or surgery to bone.5 6

The prognosis of localised and regional prostate cancer is excellent while that of metastatic prostate cancer is poor. The 5-year survival rate in patients with metastatic disease has been reported to be as low as 30%1 with a mean survival of 24 to 48 months.7 8

Evidence of the importance of SREs for survival in metastatic prostate cancer is limited. Oefelein et al9 evaluated men with prostate cancer who were prescribed androgen deprivation therapy (ADT) regardless of staging. The relative risk of skeletal fracture for mortality was 7.4. In another retrospective study, patients with bone metastasis from different primary tumours were analysed.10 In the subgroup analysis, pathological fracture increased risk of death by 20% in patients with prostate cancer although the authors failed to demonstrate statistical significance.10 A population-based cohort study demonstrated that mortality in men with metastatic prostate cancer and SREs were approximately twice that of patients with no SREs.11 Treatments for prostate cancer were, however, not recorded or analysed in the study.11

The aim of this study was to investigate the impact of SREs on survival, specifically in patients with carcinoma of the prostate with bone metastasis prescribed long-term ADT. Prognostic factors of survival in patients with SREs were also investigated.

The study period was between 1 January 2006 and 31 December 2011. Patients who were diagnosed with prostate cancer and bone metastasis and who underwent either bilateral orchiectomy or were prescribed a first dose of luteinising hormone releasing hormone analogue (LHRHa) injection during the study period at either Queen Mary Hospital or Tung Wah Hospital in Hong Kong were included. Patients were followed up until death or the last follow-up taken on 31 March 2014.

Diagnosis of carcinoma of the prostate was made following transrectal ultrasound-guided prostate biopsy, incidental histological findings of transurethral resection of a prostate specimen, biochemical diagnosis of PSA of >100 ng/mL, or other histological evidence such as bone biopsy in patients who presented with pathological fracture. Presence of bone metastasis was confirmed either by bone scan or by cross-sectional imaging such as computed tomography (CT) or magnetic resonance imaging (MRI). Patients who had evidence of bone metastases at four or more sites or visceral metastasis were regarded as having high-volume disease. Patients with medical castration were prescribed regular LHRHa injection every 3 months. Patients with underlying metabolic bone disease were excluded from study. In this study, SRE was defined in patients who developed pathological fractures, cord compression related to bone metastasis, and/or those who received irradiation or prophylactic surgery to bone metastasis.6 7 Castration-resistant prostate cancer (CRPC) was diagnosed when there were at least two consecutive rises in PSA, at least 1 week apart, with PSA of >2 ng/mL.

Data were collected from the electronic clinical management system database in the government health care system. Patients who underwent bilateral orchiectomy or received the first dose of LHRHa within the study period were shortlisted, reviewed, and then recruited as eligible patients according to the inclusion criteria. Data were collected from in-patient and out-patient records and included age at diagnosis; performance status; any bone pain at diagnosis; imaging such as bone scan, CT, and MRI; volume of metastasis; details of ADT and SREs; history of metabolic bone disease; CRPC status; treatment received for prostate cancer; and date and causes of death. Two authors (KW Wong and CF Tsang) abstracted the data and were not blinded to the outcomes.

Data were analysed using the Statistical Package for the Social Sciences (Windows version 21.0; SPSS Inc, Chicago [IL], US). The primary outcome was survival time, calculated from the date of start of ADT until death or the last follow-up. Survival was described with Kaplan-Meier curves. Univariate and multivariate analyses were performed with Cox regression model to predict prognostic factors for survival. The dependent variables were time to death (overall and cancer-specific), defined as the time from the start of ADT to death. Prognostic variables significant in the univariate analyses were entered into the multivariate Cox regression models.

A total of 119 eligible patients were identified within the study period. The mean age at prostate cancer diagnosis was 75 (range, 49-94) years. Initial ADT was by bilateral orchiectomy or LHRHa injection in 58 and 61 patients, respectively, with seven patients subsequently switched from injection to bilateral orchiectomy. The median time of follow-up was 28 (range, 1-97) months. No patient was lost to follow-up during the study.

The baseline characteristics of patients are summarised in Table 1. When stratified according to the presence of SREs, the two groups did not differ significantly in total Gleason score of prostate cancer, PSA level at the time of diagnosis, PSA nadir, Eastern Cooperative Oncology Group (ECOG) performance status, volume of metastasis, presence of bone pain at the start of ADT, or mode of ADT. Patients with SREs were slightly younger at the time of diagnosis (73.1 vs 76.3 years; P=0.04) and had a shorter mean follow-up time (28.5 vs 39.1 months; P=0.02). More patients with SREs developed CRPC when compared with those who did not have SREs (84.6% vs 65.7%; P=0.02).

The treatment received by patients with and without SREs were compared (Table 1). Only treatments received prior to development of SREs were included in Table 1 to analyse whether the baseline characteristics of treatment differed before the development of SREs. The proportion of patients prescribed chemotherapy, bicalutamide, flutamide, ketoconazole, cyproterone acetate, and calcium supplement was statistically similar for the two groups. Only two patients in each group received abiraterone and denosumab therapy. No patient received sipuleucel-T, cabazitaxel, enzalutamide, radium-223, or other novel treatment for prostate cancer throughout the study period.

Of 119 patients, 52 (43.7%) developed SREs. A total of 69 SREs were recorded—36 (69.2%) patients had one SRE, 15 (28.8%) patients had two SREs, and one patient had three SREs. Irradiation to bone for pain control accounted for 47 (68.1%) events; 14 (20.3%) events were cord compression and there were eight (11.6%) events of pathological fractures without cord compression. No patient underwent prophylactic surgery for bone metastasis. With regard to timing of SRE development, 13 (10.9%) patients had SRE as the initial presentation of metastatic prostate cancer. The overall cumulative incidence of SREs at 1 year and 5 years of diagnosis was 23.5% and 42.9%, respectively (Fig 1).

The median time required to develop CRPC status from the start of ADT was 9 months (Fig 2a). When stratified according to the presence of SREs, the median time to CRPC status from ADT initiation was significantly shorter in patients with SREs than in those without (6 vs 12 months, log-rank test, P=0.001; Fig 2b).

The actuarial overall survival (OS) and cancer-specific survival (CSS) curves are shown in Figure 3. The 5-year actuarial OS and CSS was 32% and 43%, respectively. Among men without SREs, 38 (56.7%) patients died, compared with 44 (84.6%) patients in the SRE group. When stratified according to presence of SREs (Fig 4), the median OS and CSS for patients with SREs were significantly shorter than that for patients without SREs (log-rank test: 23 vs 48 months, P=0.003 and 26 vs 97 months, P<0.001, respectively).

Various possible factors that could affect survival were analysed (Table 2a). All treatments for prostate cancer received both before and after SRE were included. Univariate analysis revealed that in terms of OS, presence of SREs (P=0.003), PSA nadir of >4 ng/mL (P<0.001), ECOG grade 2 or above (P=0.01), and calcium supplement (P=0.03) were significant risk factors. On multivariate analysis, only the presence of SREs and PSA nadir of >4 ng/mL remained statistically significant, with hazard ratio (HR) of 2.73 (95% confidence interval [CI], 1.46-5.10; P=0.002) and 3.01 (95% CI, 1.54-5.90; P=0.001), respectively. In terms of CSS, presence of SREs (P<0.001), PSA nadir of >4 ng/mL (P=0.004), and ketoconazole therapy (P=0.05) remained significant risk factors on both univariate and multivariate analyses. The HR for the presence of SREs, PSA nadir of >4 ng/mL, and ketoconazole therapy was 3.92 (95% CI, 1.87-8.23; P<0.001), 2.98 (95% CI, 1.43-6.23; P=0.004), and 2.10 (95% CI, 1.01-4.38; P=0.05), respectively.

The median survival period after occurrence of SRE was 11.5 months. A post-hoc analysis for OS and CSS after SRE revealed PSA nadir of >4 ng/mL as the only independent predictor for survival after SRE in both univariate and multivariate analyses, with HR of 10.42 (95% CI, 2.10-51.66; P=0.004) and 10.54 (95% CI, 1.94-57.28; P=0.006), respectively (Table 2b).

The importance of SREs in survival of patients with prostate cancer with different disease stage and treatments was studied10 11 12 but not specifically in patients with metastatic prostate cancer prescribed ADT. This group of patients was selected because patients with metastatic prostate cancer are at risk of developing SREs.11 In addition, ADT is the standard first-line treatment for metastatic prostate cancer.12 It has been proven to provide a clear benefit in terms of preventing SREs.13 Focusing on patients who are prescribed ADT can ensure that the effect of ADT in preventing SREs is balanced out during analysis. A study by Oefelein et al9 showed that skeletal fractures negatively correlate with OS in men with prostate cancer prescribed ADT, but it included patients with localised disease and all kinds of fracture including osteoporotic fractures. Berruti et al4 reported the incidence of skeletal complications in patients with CRPC and bone metastasis, but failed to demonstrate any difference in survival between patients with and without skeletal complications. Multivariate analysis was not performed on survival either. Daniell et al14 15 reported eight fractures in 49 patients with prostate cancer at various times following orchiectomy but did not take into account the preventive effect of ADT in SREs.13 In our study, patients with SREs had much worse OS and CSS when compared with those without SREs (23 vs 48 months and 26 vs 97 months, respectively), and remained significantly so after multivariate Cox regression analysis. To our knowledge, this is the first reported study to investigate the impact of SREs on survival in this homogeneous group of patients.

The baseline characteristics were similar between patients with or without SREs except that those with SREs were slightly younger at diagnosis (73.1 vs 76.3 years; P=0.04). This, however, does not affect data interpretation since age at diagnosis was not a significant factor in subsequent analyses for both OS and CSS. In our targeted group of patients with metastatic prostate cancer prescribed ADT, the presence of SREs was first shown to be an independent predictive factor for OS and CSS with a notable HR of 2.73 and 3.92 respectively, taking into account baseline cancer characteristics, ECOG performance status, development of CRPC status, and different treatments received. In addition, PSA nadir was found to be another predictive factor for OS and CSS. This finding has been reported in previous studies16 although most included patients who were heterogeneous in terms of clinical stage of prostate cancer. Kitagawa et al16 showed that PSA nadir of >4 ng/mL was associated with HR of 5.22 (95% CI, 2.757-9.89; P<0.001) in OS in patients with prostate cancer. The cohort, however, included patients with either locally advanced non-metastatic disease or metastatic disease. Park et al17 reported that a higher PSA nadir level correlated with shorter CSS. Similar to the previous study,16 patients with lymph node metastasis were also included. In another retrospective study,18 a high PSA nadir level was shown to be associated with shorter OS in a homogeneous group of patients with metastatic prostate cancer prescribed ADT. Nonetheless only 87 patients were included in the study. In our study, in patients who developed SREs, PSA nadir was the only predictive factor for both OS and CSS with HR of 10.42 and 10.54, respectively. This is previously unreported.

Various treatments have been proven to improve OS in patients with metastatic prostate cancer, including docetaxel,19 cabazitaxel,20 abiraterone,21 22 23 sipuleucel-T,24 and enzalutamide.25 26 Various bone-modulating agents have also been studied for patients with bone metastasis. Bisphosphonate therapy has been shown to improve bone mineral density and quality of life,27 28 and reduce the incidence of SREs in patients with metastatic CRPC in a randomised controlled trial (RCT), although there was no proven survival benefit.6 The receptor activator for nuclear factor κB ligand (RANKL) inhibitor denosumab is another bone-modulating agent proven to reduce the incidence of SREs in metastatic CRPC patients but also without survival benefit.29 30 Radium-223, a bone-seeking calcium-mimicking alpha emitter, was shown in a RCT31 to not only delay first symptomatic SRE, but also improve OS. Therefore, when investigating the incidence of SRE and survival in these groups of patients, the aforementioned treatments have to be taken into account. In our study, treatments received by patients without SREs and in patients prior to development of SREs were statistically similar (Table 1). The number of patients prescribed chemotherapy or novel hormonal agents was relatively small in our series. Sipuleucel-T, enzalutamide, and radium-223 were not available in this locality during the study period. Denosumab and abiraterone therapies were used by only two patients in each group as these medications were not subsidised by the local government and were not affordable for many patients. Docetaxel has been shown to improve bone pain and OS in a phase III RCT.19 After development of SREs, six more patients received chemotherapy in our series. All but one patient received docetaxel. The remaining patient received estramustine and etoposide before development of SREs. The fact that all patients prescribed docetaxel were in the SRE group suggests that its potential benefit in improving OS has been offset by SREs and so this is not a confounding factor in our study.

The overall prevalence of bisphosphonate therapy was low (17%). Nine patients received bisphosphonate therapy only after development of SREs. In fact, in patients receiving bisphosphonate therapy, two out of seven patients without SREs and six out of 13 patients with SREs only received one dose of bisphosphonate due to various reasons, including side-effects of the medication, affordability, and early mortality after medication. With the heterogeneous timing of start and duration of therapy, we cannot accurately comment on the benefit of bisphosphonate in our series. No further patient received RANKL inhibitor or abiraterone after development of SREs.

Since the pre-chemotherapy era, the concept of complete androgen blockade with classic hormonal manipulation by both steroidal anti-androgen, such as cyproterone acetate,32 and non-steroidal anti-androgen (such as bicalutamide,33 flutamide,34 and nilutamide35) has been widely adopted when patients develop CRPC status. This practice remains in use in this locality despite the fact that no associated survival benefit has ever been reported12 due to the side-effects and availabilities of aforementioned novel treatments for CRPC. Ketoconazole, a broad-spectrum imidazole antifungal agent, was previously the hormonal treatment of choice after anti-androgen withdrawal for complete androgen blockade.35 It works by preventing adrenal steroidogenesis with inhibition of the enzyme cytochrome P450 14 alpha-demethylase.36 Bicalutamide, flutamide, cyproterone acetate, and ketoconazole were used in our centre for hormonal manipulation. Interestingly, ketoconazole use appeared to have a deleterious effect on CSS even with multivariate Cox regression in our study. This result contradicts that of a phase III RCT35 which showed positive PSA and objective response but no survival benefit or harm. As our study was retrospective in nature, the implication of ketoconazole use is doubtful based on the results of this study and requires further evaluation.

With a median follow-up of 28 months, the incidence of SREs in men with metastatic prostate cancer was high (43.7%) and is comparable with 43.6% reported from the Danish group population-based cohort study with similar follow-up period.11 The median time to CRPC status from first ADT was 9 months, which is 5.7 months shorter than the control arm of the CHAARTED trial.37 This may be explained by the fact that the CHAARTED trial included patients prescribed ADT for less than 24 months but those with disease progression within 12 months were excluded.

We obtained local data of the natural history of metastatic prostate cancer with or without SREs and the impact of SREs on survival. A PSA nadir of >4 ng/mL was an independent poor prognostic factor for OS and CSS after development of SREs. Its clinical use in terms of predicting prognosis and patient counselling is highly feasible. Based on our results, prevention of SREs in patients with metastatic prostate cancer may translate to longer survival. Nonetheless most bone-targeting therapies, including bisphosphonate therapy and RANKL inhibitors, have failed to demonstrate survival benefit even though they prevent SREs.6 30 34 38 Radium-223 appears to hold promise as it delays symptomatic SREs by 5.8 months and improves OS by 3.8 months in metastatic prostate cancer patients.31 Further studies are needed in this field.

There are several limitations in this study. This was a retrospective study with small sample size so statistical power is limited. There are even fewer patients in post-hoc analysis. The data collected may not accurately reflect the condition of patients because the follow-up protocol was not standardised. Furthermore, the data abstraction process was not blinded. For better presentation of data, several factors such as PSA nadir, initial PSA, and age at diagnosis were analysed as categorical data. This could lead to information bias. The definition of CRPC was less stringent than that suggested from international guidelines12 because testosterone level and follow-up imaging such as bone scans were not routinely performed due to limited resources. Potential confounding factors for survival such as smoking and co-morbidity were also not included in the study and may have affected the validity of the results. The small number of patients prescribed novel treatments or bone-modulating agents did not allow a comprehensive understanding of their influence on SRE occurrence. Further prospective trials with a large cohort size are necessary.

Skeletal-related events were common in men with metastatic prostate cancer and were first shown by this study to be an independent prognostic factor of OS and CSS in patients with metastatic prostate cancer prescribed ADT. A PSA nadir of >4 ng/mL is an independent poor prognostic factor for OS and CSS following development of SREs.

1. Siegel R, Ward E, Brawley O, Jemal A. Cancer statistics, 2011: the impact of eliminating socioeconomic and racial disparities on premature cancer deaths. CA Cancer J Clin 2011;61:212-36. Crossref

2. Hong Kong Cancer Registry 2012. Available from: http://www3.ha.org.hk/cancereg/Summary%20of%20CanStat%202012.pdf. Accessed 6 Oct 2015.

3. Mullan RJ, Jacobsen SJ, Bergstralh EJ, et al. Decline in the overall incidence of regional-distant prostate cancer in Olmsted County, MN, 1980-2000. BJU Int 2005;95:951-5. Crossref

4. Berruti A, Dogliotti L, Bitossi R, et al. Incidence of skeletal complications in patients with bone metastatic prostate cancer and hormone refractory disease: predictive role of bone resorption and formation markers evaluated at baseline. J Urol 2000;164:1248-53. Crossref

5. Saad F, Gleason DM, Murray R, et al. A randomized, placebo-controlled trial of zoledronic acid in patients with hormone-refractory metastatic prostate carcinoma. J Natl Cancer Inst 2002;94:1458-68. Crossref

6. Saad F, Gleason DM, Murray R, et al. Long-term efficacy of zoledronic acid for the prevention of skeletal complications in patients with metastatic hormone-refractory prostate cancer. J Natl Cancer Inst 2004;96:879-82. Crossref

7. Galasko CS. Skeletal metastases. Clin Orthop Relat Res 1986;210:18-30. Crossref

8. Logothetis CJ, Navone NM, Lin SH. Understanding the biology of bone metastases: key to the effective treatment of prostate cancer. Clin Cancer Res 2008;14:1599-602. Crossref

9. Oefelein MG, Ricchiuti V, Conrad W, Resnick MI. Skeletal fractures negatively correlate with overall survival in men with prostate cancer. J Urol 2002;168:1005-7. Crossref

10. Saad F, Lipton A, Cook R, Chen YM, Smith M, Coleman R. Pathologic fractures correlate with reduced survival in patients with malignant bone disease. Cancer 2007;110:1860-7. Crossref

11. Nørgaard M, Jensen AØ, Jacobsen JB, Cetin K, Fryzek JP, Sørensen HT. Skeletal related events, bone metastasis and survival of prostate cancer: a population based cohort study in Denmark (1999 to 2007). J Urol 2010;184:162-7. Crossref

12. Mottet N, Bastian PJ, Bellmunt J, et al. European Association of Urology Guidelines on Prostate Cancer 2014. Available from: http://uroweb.org/wp-content/uploads/1607-Prostate-Cancer_LRV3.pdf. Accessed 6 Oct 2015.

13. Nair B, Wilt T, MacDonald R, Rutks I. Early versus deferred androgen suppression in the treatment of advanced prostatic cancer. Cochrane Database Syst Rev 2002;(1):CD003506.

14. Daniell HW. Osteoporosis after orchiectomy for prostate cancer. J Urol 1997;157:439-44. Crossref

15. Daniell HW, Dunn SR, Ferguson DW, Lomas G, Niazi Z, Stratte PT. Progressive osteoporosis during androgen deprivation therapy for prostate cancer. J Urol 2000;163:181-6. Crossref

16. Kitagawa Y, Ueno S, Izumi K, et al. Nadir prostate-specific antigen (PSA) level and time to PSA nadir following primary androgen deprivation therapy as independent prognostic factors in a Japanese large-scale prospective cohort study (J-CaP). J Cancer Res Clin Oncol 2014;140:673-9. Crossref

17. Park YH, Hwang IS, Jeong CW, Kim HH, Lee SE, Kwak C. Prostate specific antigen half-time and prostate specific antigen doubling time as predictors of response to androgen deprivation therapy for metastatic prostate cancer. J Urol 2009;181:2520-4; discussion 2525. Crossref

18. Sasaki T, Onishi T, Hoshina A. Nadir PSA level and time to PSA nadir following primary androgen deprivation therapy are the early survival predictors for prostate cancer patients with bone metastasis. Prostate Cancer Prostatic Dis 2011;14:248-52. Crossref

19. Tannock IF, de Wit R, Berry WR, et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 2004;351:1502-12. Crossref

20. de Bono JS, Oudard S, Ozguroglu M, et al. Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after docetaxel treatment: a randomised open-label trial. Lancet 2010;376:1147-54. Crossref

21. Ryan CJ, Smith MR, de Bono JS, et al. Abiraterone in metastatic prostate cancer without previous chemotherapy. N Engl J Med 2013;368:138-48. Crossref

22. de Bono JS, Logothetis CJ, Molina A, et al. Abiraterone and increased survival in metastatic prostate cancer. N Engl J Med 2011;364:1995-2005. Crossref

23. Fizazi K, Scher HI, Molina A, et al. Abiraterone acetate for treatment of metastatic castration-resistant prostate cancer: final overall survival analysis of the COU-AA-301 randomised, double-blind, placebo-controlled phase 3 study. Lancet Oncol 2012;10:983-92. Crossref

24. Kantoff PW, Higano CS, Shore ND, et al. Sipuleucel-T immunotherapy for castration-resistant prostate cancer. N Engl J Med 2010;363:411-22. Crossref

25. Beer TM, Armstrong AJ, Rathkopf DE, et al. Enzalutamide in metastatic prostate cancer before chemotherapy. N Engl J Med 2014;371:424-33. Crossref

26. Scher HI, Fizazi K, Saad F, et al. Increased survival with enzalutamide in prostate cancer after chemotherapy. N Engl J Med 2012;367:1187-97. Crossref

27. Saad F. Maintaining bone health throughout the continuum of care for prostate cancer. In: Progress in bone cancer research. Hauppauge, NY: Nova Science Publishers; 2006.

28. Smith MR. Bisphosphonates to prevent osteoporosis in men receiving androgen deprivation therapy for prostate cancer. Drugs Aging 2003;20:175-83. Crossref

29. Smith MR, Egerdie B, Hernández Toriz N, et al. Denosumab in men receiving androgen-deprivation therapy for prostate cancer. N Engl J Med 2009;361:745-55. Crossref

30. Fizazi K, Carducci M, Smith M, et al. Denosumab versus zoledronic acid for treatment of bone metastases in men with castration-resistant prostate cancer: a randomised, double-blind study. Lancet 2011;377:813-22. Crossref

31. Parker C, Nilsson S, Heinrich D, et al. Alpha emitter radium-223 and survival in metastatic prostate cancer. N Engl J Med 2013;369:213-23. Crossref

32. Goldenberg SL, Bruchovsky N. Use of cyproterone acetate in prostate cancer. Urol Clin North Am 1991;18:111-22.

33. Scher HI, Liebertz C, Kelly WK, et al. Bicalutamide for advanced prostate cancer: the natural versus treated history of disease. J Clin Oncol 1997;15:2928-38.

34. Crawford ED, Eisenberger MA, McLeod DG, et al. A controlled trial of leuprolide with and without flutamide in prostatic carcinoma. N Engl J Med 1989;321:419-24. Crossref

35. Small EJ, Halabi S, Dawson NA, et al. Antiandrogen withdrawal alone or in combination with ketoconazole in androgen-independent prostate cancer patients: a phase III trial (CALGB 9583). J Clin Oncol 2004;22:1025-33. Crossref

36. Loose DS, Kan PB, Hirst MA, Marcus RA, Feldman D. Ketoconazole blocks adrenal steroidogenesis by inhibiting cytochrome P450-dependent enzymes. J Clin Invest 1983;71:1495-9. Crossref

37. Sweeney CJ, Chen YH, Carducci M, et al. Chemohormonal therapy in metastatic hormone-sensitive prostate cancer. N Engl J Med 2015;373:737-46. Crossref

38. Dole EJ, Holdsworth MT. Nilutamide: an antiandrogen for the treatment of prostate cancer. Ann Pharmacother 1997;31:65-75.