Introduction

Surgical male circumcision(MC) is among the oldest surgical procedures worldwide[1], motivated by various cultural, religious, hygienic and medical reasons [2]. Since the 1960s, MC has been one of the most effective strategies to prevent sexually transmitted infections and to address complications from sexual intercourse[1,3]. More recent observational and randomized controlled trials have shown that MC reduces the risk of HIV infection among adult males by at least 50% [4,5]. The vast majority of all new HIV infections occur in sub-Saharan Africa and 70% are due to heterosexual intercourse[4]. In this region, poverty, among other factors, limits access to HIV prevention and treatment. As a result, WHO/UNAIDS recommends universal MC as a population-level HIV prevention strategy in countries with high HIV prevalence driven by heterosexual transmission and low male circumcision prevalence[4,6].

To have an impact on HIV transmission, WHO/UNAIDS set a target of 20 million men for male circumcision in Africa by 2015[7]. Although surgery has been the most commonly used MC procedure in adults, risks include bleeding, infections, mortality from anesthesia, and in rare cases, complications that may lead to penile damage or sexual impotence[1,4]. Surgical MC is a relatively time-intensive procedure performed in a sterile environment by a physician using many instruments, with intensive care during the wound period[3,4].For these reasons, surgical MC is not ideal for mass intervention in adult men in Africa due to limited resources and few equipped health facilities and physicians[8].

In Rwanda, 3% of adults are infected with HIV, and only 13% of Rwandan men were circumcised in 2010[9]. In 2009, the Rwanda Ministry of Health set the target of 2 million adult male circumcisions by the end of 2012[10].With only 1 physician per 16,000 inhabitants[11], the Rwandan government considered alternatives to surgical MC including the Shang Ring device [12] and the circular cutter with stapled anastomosis [13].The Ministry of Health opted for PrePex, a quick, non-surgical procedure that does not require highly qualified staff or a sterile environment [14]. With three clinical trials in Rwanda, the procedure was pre-qualified in March 2012 by the WHO[10].Rwanda was the first country to adopt PrePex in the public health setting on a mass scale.

There are several benefits to PrePex including the ability to shift the MC procedures to less-trained staff, which makes the procedure cost-effective and scalable[15], even in high resource settings[16].Further, the minimal invasiveness and risk make the procedure acceptable to beneficiaries[17]. However, there is little research describing the implementation of PrePex in a real world setting, outside of a research study environment. Rwanda Military Hospital (RMH), a referral hospital in the country’s capital city, Kigali, which serves military members and civilians (80% of patients) was the first hospital in the country to offer PrePex MC procedure starting in 2011. The goal of this paper is to describe the follow-up and outcomes of the first clients who underwent PrePex MC procedure at Rwanda Military Hospital (RMH).

Methods

Results

Between January 2011 and December 2012, 639 men were screened for MC, 66 men did not meet inclusion criteria, and 570 men received MC with PrePex at RMH. Among these clients, 167 (29.3%) were formally discharged and 403 (70.7%) defaulted before formal discharge. Most clients were between 21 and 24 years old (57.7%) or 25 and 29 years (27.7%), and nearly all clients were circumcised by a non-physician operator (97.0% and 96.3%, respectively, among formally discharged and defaulters). Most men were HIV negative (99.4% among formally discharged and 98.8% among defaulters) and had no evidence of urological problems that could delay their healing and formal discharge (97.0% among formally discharged and 98.3% among defaulters). Very few men reported pain either before the device removal (0.5%) or after the device removal (0.7%). Among all clients, 37.5% were fitted with the device size B (28 mm), although, there was no statistical significant between the device size and defaulting status (p = 0.725). Pain before (1.2% versus 0.3%) and after the device removal (0.6% versus 0.7%) were similar, but more clients were missing pain data among defaulters versus clients who were formally discharged (p<0.001). The median follow-up time from device application date was seven weeks (IQR: 5, 9) and six weeks (IQR: 1.8) respectively among formally discharged and defaulter clients (p = 0.001) (Table 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Bivariate associations between discharge status and socio-demographic and clinical characteristics at Rwanda Military Hospital, 2011 (N = 570).

https://doi.org/10.1371/journal.pone.0138287.t001

Few adverse events were recorded (four cases of diffuse edema, four of bleeding, five of wound infection, three of productive exudate and eleven others), of which20 (4.9%) were among defaulters and seven (4.2%) were among formally discharged (Table 2). Most clients healed between six and eight weeks (68.3%), very few (6.6%) required more than eight weeks to heal; and seven adverse events occurred after the sixth week (Table 3).Most clients defaulted between six and eight weeks (42.2%), few defaulted between nine and ten weeks (0.7%); and four adverse events occurred between the second and fourth weeks, and 16 after the sixth week (Table 4).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Number of adverse events among those who defaulted and remained.

https://doi.org/10.1371/journal.pone.0138287.t002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Time and cumulative probability to formal discharge/healing(N = 167).

https://doi.org/10.1371/journal.pone.0138287.t003

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. Time and cumulative probability to defaulting (N = 403).

https://doi.org/10.1371/journal.pone.0138287.t004

Discussion

The results of this real-world scale-up of PrePex MC at RMH are promising for the scale-up of PrePex in other similar settings. In the first two years, 570 adult men were circumcised with PrePex, primarily by non-physician staff. The few instances of adverse events and the recorded healing times were consistent with larger clinical studies on PrePex where the median time for healing was five to six weeks [18]. PrePex takes on average one to two weeks longer to heal than surgical MC[19,20] or Shang ring[21].

The biggest challenge in this setting was the high rate of defaulting; a majority of patients discontinued weekly follow-up visits before being formally discharged from the program. The high rate of default during and after the sixth week might be because clients perceived that they were healing properly and they did not feel a need to return for clinical assessments. We suspect that a client with a significant problem would have come back for clinical follow-up. Given the high rate of default before formal discharge, RMH is now using a protocol with less frequent follow-up visits to improve adherence without compromising clinical outcomes.

The only socio-demographic or clinical difference between the groups was the amount of missing data in reported pain before and after device removal. One explanation for this difference in missing data is that clients who defaulted tended to be the same types of clients who do not respond to all questions during removal of the PrePex device. Another explanation is that clients who were not asked about pain during device removal felt less rapport with clinical staff and were more likely to default, though ongoing supervision of clinical staff throughout the study makes this explanation unlikely.

The findings about adverse events among defaulters suggest that there were not substantial differences in number, type, and timing of adverse events and experiences of pain between the two groups which might drive someone to default. Other studies assessing the efficacy of PrePex procedure in a more formal study environment in Uganda and Rwanda also found few adverse events among clients[14,17,18,22] and it is reassuring to see this continue in a real-world setting. Given the low rate of adverse events and large amount of defaulting, implementers should consider fewer follow-up visits.

There are some limitations to this study. HIV, urological problems and adverse events were all rare, and our sample may have been too small to detect differences between the defaulter and non-defaulter groups. However, there is no indication of a difference between the groups and we hypothesize that individuals are defaulting before formal discharge because they are healing well. This should be studied more in the future. Additionally, although we collected information about type of adverse event, information about the severity of adverse events was not collected and should be collected in future studies. Finally, RMH is a national referral hospital with different services and infrastructure compared to district hospitals and health centers where this procedure is mostly likely to be scaled. However, as PrePex procedure is performed in a clean environment, with non-physicians and without anesthesia, we expect that outcomes and few complications observed in this study would be replicated in other clinical settings in Rwanda and in sub-Saharan Africa.

Rwanda has been successful in implementing PrePex as a non-surgical MC procedure, a strategy to supplement surgical MC to achieve the Ministry of Health target of two million adult male circumcisions by 2012.However, the number of men circumcised in this two years period at RMH is small compared to the two million targeted men; as of 2014, PrePex is only implemented at RMH health facilities in Rwanda and during army weeks and RMH outreach activities. Given these promising results, and that the procedure can be performed in a non-sterile environment with a non-physician with very few adverse events, we strongly encourage scale up of PrePex MC at all district hospitals in Rwanda. However, more evaluation of post PrePex follow-up protocols should be conducted to encourage clinical monitoring of MC clients until full documented healing.

Acknowledgments

Disclaimer: This document’s contents are solely the responsibility of the authors and do not necessarily represent the official view of the donors.

The paper was developed as part of a Junior Faculty Seminar Series facilitated by authors BHG and DRT at the UR-CMHS-SPH under the leadership of the School of Public Health. We acknowledge the support of the Commandant of Rwanda Military Hospital for his acceptance to use RMH client records for this study. Finally, we thank the UR-CMHS-SPH data manager, who helped in data entry activities.