Is Hba1c level of diabetic patients associated with penile prosthesis implantation infections?

Abstract

Background and objectives: Diabetes mellitus is an important risk factor for erectile dysfunction (ED). Penile prosthesis implantation surgery is the final solution for diabetic patients with ED, but infections thereof are still a serious risk factor. While some studies suggest that most infections associated with penile prosthesis implantation are associated to high glycated hemoglobin (HbA1c) levels, other research did support such relationship.

Materials and methods: The current study assessed retrospectively, the association between HbA1c level and penile prosthesis surgery infection. We retrieved and reviewed the records of 300 diabetic patients who had penile prosthesis surgery at our Institution (January 2012–November 2016). Patients’ mean age was 55.26 ± 10.9 years (31% patients were <50 years of age), and mean HbA1c was 7.60 ± 1.90%.

Results: Infection rate among diabetics was 0.67%. Prevalence of prosthesis infection among patients with HbA1c ≤ 9% was 0.9%, compared with 0% among patients with HbA1c > 9%. Prosthesis infection risk did not significantly increase with higher HbA1c levels, with no meaningful difference in the median or mean level of HbA1c in the infected and non-infected diabetic patients.

Conclusion: Findings do not support the use of HbA1c values among diabetic patients who are candidates for penile prosthesis implantation surgery in order to identify and exclude those who might be prone to increased risk of prosthesis infections. Future studies would benefit from larger sample sizes in order to support or refute our findings.

Keywords:

• Diabetes mellitus
• penile prosthesis
• hemoglobin A
• glycosylated
• infection

Introduction

Types 1 and 2 diabetes mellitus (DM) are both associated with increased erectile dysfunction (ED) risk [1], to the extent that treated diabetic men are three times more likely to develop ED compared with nondiabetic men [2]. Worldwide, ≥50% of diabetic men suffer from ED [1]. Such association between DM and ED is multifactorial and attributed to vasculo/neuropathies [3], visceral adiposity, insulin resistance, and hypogonadism [4], and hyperglycemia, a main determinant of micro/vascular diabetic complications, contributes to the pathogenic mechanisms of sexual dysfunction [5]. Moreover, diabetics often present with hypertension, overweight/obesity, metabolic syndrome, atherogenic dyslipidemia, and cigarette smoking, lower Testosterone and Vitamin D levels [6], which are all risk factors for sexual dysfunction [4]. In addition, diabetic men with ED are less responsive to phosphodiesterase type 5 inhibitors when compared with nondiabetic men with ED, probably due to vascular smooth muscle and fibrotic changes [7].

Penile prostheses (PPs) are a well-established treatment for severe ED and a final option when other ED treatment modalities fail [3], but postoperative infections due to genitourinary prosthetic surgery remain a serious complication [8] and rates of initial implant placement vary between 1 and 3%, but have decreased due to e.g. antibiotic-coated devices and chlorhexidine–alcohol skin preparation [8]. Although ≈30% of infected PPs have negative cultures [9], less virulent infections are mostly caused by Staphylococcus epidermidis, whilst more aggressive infections are due to Pseudomonas aeruginosa, methicillin-resistant Staphylococcus aureus, Serratia marcescens, Enterococcus, and Proteus mirabilis [8]. Recently, Candida (11.1%), anaerobes (10.5%) and methicillin-resistant Staphylococcus aureus (9.2%) comprised ≈30% of 153 positive cultures; and multi-organism occurred in 25% of positive cultures [9].

Red blood cells in the human body survive for 120 days before renewal; hence measuring glycated hemoglobin (HbA1c) reflects the average blood glucose levels over that duration, providing a useful longer-term assessment of blood glucose control. Tight glycemic control (maintaining an HbA1c concentration <7%) is recommended for diabetic adults to minimize long-term microvascular complications [10].

The adult Qatari population exhibits a moderately high (16.6%) DM prevalence, with 10.7% diagnosed, and 5.9% newly diagnosed DM [11]. Whilst most patients in Qatar who require PP implantation (PPI) have uncontrolled DM (HbA1c levels >9%), very limited evidence exists on the HbA1c level cut off value that is associated with safe PPI and with minimal complications among diabetics [12–14]. Such lack of clear-cut off HbA1c value for safe PPI with minimal complications represents a serious gap in literature, and the current thin evidence base suggests the need for research about whether diabetic patients' HbA1c values are associated with increased PPI infection. Such information is particularly important in countries with large numbers of uncontrolled diabetic patients who need and undertake PPI, e.g. Qatar. The current study had been assessed, among 300 Qatari diabetic men with established severe ED, to look at the association between HbA1c and the outcome of their PPI in terms of infection and complications.

Methods

Ethics, sample and setting

This study was undertaken at the Andrology Unit of Hamad General Hospital, Doha, Qatar. The Medical Research Centre (IRB) approved the study (Protocol # 16394/16). Search for data was retrieved from the medical records of all patients who underwent PPI surgery (January 2012–November 2016), and conducted a retrospective chart analysis. Approximately 90% of the ED cases across Qatar are treated at our Andrology Unit, hence our sample is quite representative of the diabetic patients with ED in Qatar. The main eligibility criterion of the current work was diabetic patients who underwent PPI, and all those eligible were included in the study.

Procedures

All surgeries were performed by the same two senior surgeons who implant ≥80 PP per year. All patients had similar preoperative preparation without regard to diabetic status, control or HbA1c level. Patients had either general or spinal anesthesia based on patients’ preferences and/or the preference and experience of the anesthetist. Our routine departmental PPI surgery practice is that the assigned nurse shaves the external genitalia hair by clipper just before the surgery in operating room. Then, a nurse scrubs the external genitalia for 10 min with betadine solution diluted with saline. After the corporal dilatation is undertaken and reservoir space is created (when 3-piece PP are used), and just before the insertion of components of PP, gloves of the surgery team are changed and the prosthesis placed over a new sterile drape. All surgeries in this study were conducted in supine position via the penoscrotal approach, and reservoirs were placed into the extra-peritoneal space by blind puncture through the transversalis fascia.

Medical regime: antibiotics and pain control

Patients with normal renal function: on the day of surgery, patients received vancomycin 500 mg/i.v/two times/day and gentamycin 80 mg/i.v/three times/day infusion, where the first doses of these two antibiotics were given one hour prior to the incision of skin. During the surgery, ten vancomycin 500 mg and ten gentamycin 80 mg ampoules mixed with 500 ml saline and were used to irrigate the corporal bodies. Starting from the second postoperative day, patients received ceftriaxone 2 gm/i.v/once a day for 7 days and ciprofloxacin 500 mg/p.o/two times daily for 14 days. For pain control, we prescribed paracetamol 500 mg/p.o/three times/day and tramadol 50 mg/p.o/two times/day for 10 days.

Patients with impaired renal function: on the day of surgery, patients received ceftriaxone 2 gm/i.v/once a day and clindamycin 600 mg/i.v/two times/day infusion, where the first doses of these two antibiotics medicines were given one hour prior to the incision of skin. During the surgery, ten rifampicin 600 mg and ten gentamycin 80 mg ampoules mixed with 500 ml saline to irrigate the corporal bodies. Starting from the second day, patients received, ceftriaxone 2 gm/i.v/once a day for 7 days and ciprofloxacin 500 mg/p.o/two times daily for 14 days. For pain control, we prescribed paracetamol 500 mg/p.o/three times/day and tramadol 50 mg/p.o/two times/day for 10 days.

Statistical analysis

All statistical analyses performed using SPSS 22.0 (SPSS Inc. Chicago, IL) and Epi-info (Centers for Disease Control and Prevention, Atlanta, GA) statistical software, with p set at <.05. Quantitative data were expressed as frequency and percentage or mean ± SD and range. Descriptive statistics summarized the sample’s demographic and other clinical characteristics. Associations between two or more qualitative variables were assessed using chi-square test and Fisher exact test as appropriate. Quantitative data between two independent groups were analyzed using unpaired “t” and Mann–Whitney U tests as appropriate. Relationships between two quantitative variables (e.g. age and HbA1c, etc.) were examined using Pearson’s or Spearman’s correlation coefficients.

Results

A total of 300 PP patient records were retrieved during the study period (January 2012–November 2016), which included all DM patients who underwent different prosthesis types. Participants’ demographic, clinical, prosthesis and other related characteristics are presented in Table 1. Mean age was 55.26 ± 10.89 years old (31% patients are <50 years of age), mean corporal size was 19.71 ± 1.73 cm and mean HbA1c was 7.60 ± 1.90%. About 34% (101/300) of the patients had HbA1c values ≤6.5%. Post radical prostatectomy PPI was performed in 2.7% (8/300) of the cases.

Table 1. Selected characteristics of the sample (N = 300).

Characteristic	Value
Age, years, mean ± SD (range)	55.26 ± 10.89 (23–79)
Diabetes mellitus
HbA1c, %, mean ± SD (range)	7.6 ± 1.9 (4.1–13.2)
HbA1c (>9%), %, n/N	24.0 (72/300)
Presentation % (n/N)
Diabetes mellitus	81.0 (243/300)
Hypertension	54.3 (163/300)
Peyronie’s disease	39.3 (118/300)
Anti-coagulant use	49.5 (147/300)
Surgical characteristics
Mean operative time, mean ± SD (minute)
Malleable prosthesis	40 ± 7
2-piece prosthesis	45 ± 5
3-piece prosthesis	55 ± 8
Mean corporal size, mean ± SD (range) cm	19.71 ± 1.73 (14–25)
Post-radical prostatectomy	2.7 (8/300)
Penile prosthesis^a n (%)
Ambicor (Boston Scientific)	39 (13.0)
CX700 (Boston Scientific)	167 (55.7)
Genesis, Malleable (Coloplast Ltd.)	49 (16.3)
Titan OTR (Coloplast Ltd.)	45 (15.0)
Complications % (n/N)
Skin infection	2.3 (7/300)
Prosthesis infection	0.7 (2/300)
Erosion	0.7 (2/300)
Urethral injury	0.3 (1/300)
Device failure	2% (6/300)

^a Model (manufacturer).

The prosthesis infection rate was 0.67% (2/300) (95% CI: 0.18–2.40). The prevalence of prosthesis infection among patients with HbA1c ≤ 9% was 0.9% (2/228) compared with 0% (0/72) among those with HbA1c > 9% (χ² = 0.64, p = .998, which not significant). Although skin infection was higher among patients with HbA1c ≤ 9% compared to those with HbA1c > 9%, the difference was not significant (7/228, 3.1% vs. 0/72, 0%, p = .203, which statistically not significant).

Prosthesis infection patients were slightly younger than those without prosthesis infection (51.5 ± 2.12 vs. 55.29 ± 10.89 years, without significance); although skin infection patients were slightly older 57.0 ± 11.31 than those without skin infection (57.0 ± 11.31 vs. 55.2 ± 10.86 years, not significant). Mean HbA1c in the prosthesis infection group was slightly lower compared to those without prosthesis infection (7.0 ± 0.14 vs. 7.6 ± 1.9, not significant). Similarly, mean HbA1c in skin infection patients was slightly lower than those without skin infection (7.23 ± 0.91 vs. 7.61 ± 1.91, not significant).

Among the 24% of our 300 patients (72/300) with HbA1c values >9%, the correlation between age and HbA1c was weak and not significant (Pearson r =0.034; p = .557). Infection of prosthesis (itself or its components) did not exhibit significant associations with the prosthesis type, Peyronie's disease or DM (Fisher exact and Yates corrected Chi-Square statistical tests, data not presented). However, Prosthesis types, particularly malleable (Genesis) (4/49, 8.2%) and OTR (3/45, 6.7%) had significantly higher skin infection rates compared to Ambicor (0%) and CX700 (0%) prosthesis types (p < .05, data not presented). Patients with erosion had significantly higher prosthesis infection compared to those with no erosion (2/2, 100% vs. 0/298, 0%; p < .0001, data not presented); and the erosion group had significantly higher skin infection compared to those without erosion (1/2 or 50% vs. 6/298 or 2%; p = .046). Peyronie’s disease and DM were not associated with skin infection.

Discussion

The current study is a retrospective chart review of infections associated with PP surgery among diabetic patients in Qatar. Long-term chronic conditions e.g. DM [5] or hypogonadism [6] is a risk factor for ED, where treated diabetic males had a 28% age-adjusted prevalence of ED, three times higher than the ED prevalence observed in the entire study sample (10%) [2]. Others have similarly reported that diabetic etiology of ED comprised 20% of penile-implant patients [15].

Qatar ranks fifth globally in terms of the prevalence of diabetic people aged 20–79 years, 16% of the population is diabetic, and Qatar’s ED prevalence rate is high, where recent data indicate that 172 out of 209 diabetic men (82.2%) reported ED [16,17]. In terms of management, most diabetic men either develop resistance to first and second line ED treatments or refuse those treatments, and become candidates for PPI surgeries.

In the current study, we analyzed records, which included all our diabetic patients who underwent PP surgery at our institution for different types of prostheses. Diabetes is a risk factor for surgical infection generally, however the evidence is inconsistent as to whether HbA1c level can predict an increased incidence of PP infection [12–14,18,19].

We observed that, among diabetic patients, there were two PP infections, and the PP infections were not statistically associated with patients’ HbA1c levels. Prosthesis infection rates in diabetic and non-diabetic patients reported in the literature ranged from 0.9%-5.5% [12,20,21] for initial implant placement. Our findings indicate that 0.7% of prosthesis infection rate among diabetics was lower than that reported in most studies [20,21]. PP surgery research shows some contradictions. Some studies reported an increased prosthesis infection risk with HbA1c > 11.5%, where 31% of poorly controlled diabetic men had infection, in contrast to 5% infection among adequately controlled diabetic men [12]. Others authors found that diabetics and nondiabetics had equal incidence of PPI infections [22]. Other papers found that HbA1c levels, glucose levels, and insulin dependence were not predictive of the development of postoperative infection in general [13]. We found PP infections were not statistically associated with patients’ HbA1c levels (HbA1c ≤ 9% vs. HbA1c > 9%).

Less infection rate in our cohort, could collectively contribute to preventing potential infections e.g. surgeons’ experience in terms of: prevention of infection, surgical techniques, high volume surgeons and low operative time, and medical management focused on PP infection prevention and control. As for surgeons’ experience in terms of prevention of infection, a successful PPI requires a methodical approach, safety checks, duration of procedure, and efficiency in order to minimize infection possibilities, as the first PPI attempt has the best chance for a successful outcome, with each successive surgery associated with higher infection/complication rate [23]. All the surgeries included in our current report were performed by the same two senior prosthetic surgeons who were thorough in measures to prevent infection, and careful in adopting a systematic approach to the management of the patient in order to deal with patient- or surgery-related risk factors for PP infection. The surgical experience showed to be a key to minimize risk of such infections [24]. Shaving the external genitalia hair by clipper just before the surgery in operating room, scrubbing the external genitalia for 10 min with betadine solution diluted with saline, glove change after the corporal dilatation, and the placing the prosthesis over new sterile drape. Certainly, meticulous technique and experience are important in PP surgery [25]. These features could have enhanced our observed low infection rate.

As for surgeon experience in terms of the surgical techniques, better PP outcomes were achieved with first PP when implanted by higher volume implanters [26]. Our high volume surgeons support what others had observed: that 75% of patients were treated by surgeons below the highest quartile of annual case volume (≤31 cases per year), and that these patients were 2.1–2.5 times more likely to require reoperation for IPP infection than patients treated by surgeons in the highest quartile of annual case volume [21]. Such aspects could have boosted our observed low infection rate. PPI operation time is between 40 and 55 min for malleable, 2-piece and 3-piece PP surgeries respectively. We did not find literature that reported on their operative timing, so we are unable to compare our findings with other centers. Nevertheless, we consider such operative efficiency a positive contributor to our low infection rates; quicker surgeries tend to have less need for revision for infection [25].

As for medical management and infection prevention and control, the American Urological Association implant guidelines recommend an aminoglycoside with vancomycin or a first- or second-generation cephalosporin for 24 h [27], administered at least 1 h before incision to ensure adequate tissue levels. In addition to prophylactic antibiotics, evidence from diabetic PP recipients show the use of an antibiotic-impregnated PP can decrease revisions due to infection [28]. We follow such regime (aminoglycoside with vancomycin) and continue with oral quinolones and i.v. third-generation cephalosporin except in patients with renal failure.

Across >6000 DM patients, antibiotic-impregnated PP exhibited a >1.5-fold difference, where 1.5% had revisions due to infection vs. 4.2% of those receiving a non-impregnated PP [29]. Reoperation rate for infection was 4.2% before the routine use of antibiotic impregnated implants and decreased to 1.5% with the use of antibiotic coated prostheses [21]. As in different centers, [21,28] we use either ready-to-use impregnated PP, or we impregnate PP with vancomycin or gentamycin antibiotics in the operating room before insertion. Additionally, we use antibiotics (vancomycin and gentamycin) mixed with 500 ml saline to irrigate the corporal bodies and over the incision site. All these measures could have contributed to our observed low infection rate.

This study has limitations. On the one hand, it is retrospective in nature and has not the profound significance a prospective or randomized controlled studies. On the other hand, the comparisons we undertook exhibited non-significant relationships, possibly due to the very low infection rates we observed in our sample, lower than those reported in the literature. This suggests that a larger scale studies are necessary define outcomes.

Acknowledgements

We would like to acknowledge the careful work of Dr. Prem Chandra for his assistance with the statistics used in this study and also our technician Mr. Ahmed Sandly for his help while reviewing medical records.

Disclosure statement

No conflict of interest was declared by the authors.

Additional information

Funding

A grant from the Hamad Medical Corporation primarily supported this research [Protocol # 16394/16].