Optimizing the detection of subclinical inguinal hernias in men undergoing open radical retropubic prostatectomy

Authors


Herbert Lepor*, Department of Urology, NYU School of Medicine, 150 East 32nd Street – 2nd Floor, New York, NY 10016, USA.
e-mail: herbert.lepor@med.nyu.edu

Abstract

Study Type – Diagnostic (exploratory cohort)
Level of Evidence 2b

OBJECTIVE

To evaluate the role of magnetic resonance imaging (MRI) and up-standing ultrasonography (USUS) for the detection of inguinal hernias (IHs) before open radical retropubic prostatectomy (ORRP) and to assess factors associated with the presence of IHs in these men.

PATIENTS AND METHODS

From 1 July 2007 to 1 February 2009, 164 men underwent preoperative evaluation for ORRP by one surgeon. Of these men 113 (69%) were evaluated for IHs by physical examination (PE), USUS and MRI. In all 226 groins were examined. Any IH diagnosed by any method was considered a true positive. The sensitivity, negative predictive value (NPV), and concordance were calculated for the three diagnostic methods. Possible predictive factors of IHs were evaluated.

RESULTS

Of the 226 groins evaluated, 72 (32%) IHs were diagnosed. USUS had the greatest sensitivity (69.4%) and the highest NPV (87.5%). MRI had fair agreement with PE and USUS, while USUS and PE had moderate agreement with each other. No factor was associated with an increased likelihood of preoperative diagnosis of IH.

CONCLUSIONS

This study was limited by the lack of a reference standard to diagnose IH. USUS was the most sensitive method for the detection of IH. We recommend that all men undergoing ORRP should be evaluated for IHs by PE and at least one imaging method and that IHs be repaired at the time of ORRP, obviating the need for a second surgical procedure.

Abbreviations
IH(R)

inguinal hernia (repair)

ORRP

open radical retropubic prostatectomy

NYULMC

New York University Langone Medical Center

(US)US

(up-standing) ultrasonography/ultrasound

PE

physical examination

BMI

body mass index

OLUTS

obstructive LUTS

NPV

negative predictive value

INTRODUCTION

Inguinal hernias (IHs) develop in 8.6–23.9% of men undergoing open radical retropubic prostatectomy (ORRP) [1–5]. This high rate of IHs after ORRP led to the conclusion that the surgical procedure was the direct cause of these IHs. In none of these reports was an effort made to diagnose IHs before or during ORRP and repair them. It is reasonable to postulate that men with localized prostate cancer undergoing RP have the same risk of having an IH as age-matched men in the general population. The lifetime prevalence rate of IHs for men aged 45–54 years, 55–64 years and 65–74 years has been estimated as 28.0%, 34.5%, and 39.7%, respectively [6]. Given the high prevalence rates for men aged >45 years in the general community, it is reasonable to assume that many of the IHs that develop after ORRP existed before surgical intervention. Detecting these IHs preoperatively provides the opportunity to perform a preperitoneal IH repair (IHR) at the time of RP [7,8], thereby obviating the need for a second surgical procedure.

Lepor et al. [9] reported that 12.7% of surgical candidates would have an IH diagnosed by a simple examination of the inguinal canal in the up-standing position during a Valsalva manoeuvre. All of these men diagnosed with an IH underwent a preperitoneal IHR at the time of ORRP. In that series, 8% of men developed an IH 12–60 months after surgery. Nielsen et al. [10] reported that up to 33% of men had an IH when the inguinal ring and inguinal floor were examined intraoperatively. While no IH repaired at the time of ORRP recurred, new IHs developed in 5.3% of cases, with 3.1% occurring at the side of the repair. These experiences suggest that some pre-existing IHs may not be detected by physical examination (PE) or intraoperative inspection of the inguinal floor and canal or that ORRP in fact is the direct cause of the IH.

In our prior series, routine body imaging using computerized tomography failed to diagnose the most of IHs due to the fact the procedure was performed supine without a Valsalva manoeuvre [9]. The objective of the present study was to determine the sensitivity and negative predictive value (NPV) of several diagnostic imaging methods performed with Valsalva in the up-standing position (when possible) for the preoperative detection of IHs, as well as to ascertain those factors predictive of the diagnosis of IH before ORRP.

PATIENTS AND METHODS

From 1 July 2007 to 1 February 2009, 164 men with clinically localized prostate cancer underwent a comprehensive preoperative assessment to identify IHs before ORRP. All of these men signed informed consent to participate in the New York University longitudinal outcomes database. Approval was also obtained from the New York University Langone Medical Center (NYULMC) Institutional Review Board to retrospectively gain access to the inguinal MRI and ultrasound (US) images and reports. The comprehensive assessment for the preoperative diagnosis of IHs included PE and bilateral inguinal imaging using MRI and US. The PE was performed by one surgeon (H.L.) who performed all ORRPs. The inguinal canals were palpated while patients performed a Valsalva manoeuvre in the up-standing position.

US of the bilateral inguinal canals was carried out with the patient performing a Valsalva manoeuvre while in the up-standing position. Up-standing US (USUS) was performed and interpreted by one radiologist (S.T.) using a Sequoia machine and a 10-MHz linear transducer. An examination was considered positive when the inguinal canal anteroposterior diameter was >10 mm with Valsalva (Fig. 1).

Figure 1.

USUS examination for IH: US image a is an inguinal canal before Valsalva manoeuvre. US image b is an inguinal canal containing bowel protruding after Valsalva manoeuvre.

MRI of the groins was performed during prostate MRI using 1.5 T clinical systems (Siemens Medical Solutions, Erlangen, Germany) using pelvic phased-array coils. A HASTE (Half Fourier Acquisition Single shot Turbo spin Echo) sequence was performed in the axial and coronal planes centred on the femoral vessels, at baseline and after a Valsalva manoeuvre (with the patient supine). All MRI images were interpreted by one expert body MRI radiologist (B.T.). An examination of the inguinal canals by MRI was considered positive if fat and/or bowel passed through an inguinal canal at rest and/or on Valsalva (Fig. 2).

Figure 2.

MRI examination for IH: Coronal (a) and axial (c) MRI views of the inguinal canals before Valsalva manoeuvre show fat containing left inguinal hernia (arrows). Coronal (b) and axial (d) imaging during Valsalva manoeuvre confirms the presence of an IH with small bowel protruding into the left inguinal canal (arrows).

The NYULMC longitudinal database prospectively captures relevant preoperative data related to potential risk factors for IHs including age, body mass index (BMI), severity of obstructive LUTS (OLUTS), prior IHR, history of hiatal hernia, smoking history, and prostate volume. OLUTS severity was determined from the responses to the four questions from the AUA symptom score pertaining to obstructive symptoms, including incomplete emptying, decreased calibre of stream, intermittency of stream, and straining.

As there was no ‘gold standard’ for the diagnosis of an IH, it was assumed that an IH observed by PE, MRI or USUS represented a true IH. The sensitivity and NPV of the three different methods for the diagnosis IHs were calculated based on this assumption.

A univariate analysis was used to determine those factors predictive of the preoperative detection of IHs.

RESULTS

Overall, IHs were diagnosed in 54 (48%) of the 113 men who underwent a comprehensive assessment to detect IHs before ORRP. In all, 18 (33%) of these men were diagnosed with bilateral IHs, for a total of 72 IHs of 226 groins detected before ORRP. The baseline characteristics of the entire cohort and those evaluable men are shown in Table 1. Features of the evaluable men and total series were generally similar. There was a statistically significant difference between the PSA levels of men in the two groups; however, this difference was not clinically significant.

Table 1.  Baseline characteristics
VariableTotal seriesEvaluable menP
  • *

    Mann–Whitney U-test;

  • §

    §Fisher exact test; IQR, interquartile range.

Number of men164 113 
Median (IQR):   
 Age, years61.0 (54.3, 65.0)60.0 (53.0, 66.0)0.56*
 BMI, kg/m226.5 (24.3, 29.3)26.2 (23.9, 29.1)0.10*
 Serum PSA level, ng/dL4.7 (3.7, 6.7)4.5 (3.6, 6.0)<0.001*
 AUA symptom score5.5 (2.0, 10.0)6.0 (2.0, 10.5)0.99*
N (%):   
 Clinical stage:  0.51§
  T1 116 (70.7)78 (69.0) 
  T246 (28.0)34 (31.0) 
  T32 (1.2)1 (0.9) 
 Pathological stage:  0.59§
  pT2109 (66.9)73 (65.2) 
  pT354 (33.1)39 (34.8) 
 Gleason score:  0.30§
  ≤675 (45.7)53 (46.9) 
  766 (40.2)41 (36.3) 
  ≥823 (14.0)19 (16.8) 
 Ethnicity:  0.055§
  Caucasian145 (89.5)103 (92.8) 
  Other17 (10.5)8 (7.2) 

A univariate analysis was used to determine which preoperative variable(s) predicts the presence of an IH (Table 2). None of the factors associated with the aggressiveness of the disease (serum PSA level, Gleason score, clinical stage, or pathological stage) or any factors suspected to have an association with IH (age, BMI, OLUTS, history of hiatal hernia, smoking history or prostate volume) significantly predicted the preoperative detection of an IH. A history of prior IHR approached statistical significance (P= 0.054).

Table 2. 
Factors predicting detection of IHs: univariate analysis
FactorOdds ratio (95% CI)P
Age1.04 (0.99–1.09)0.13
BMI0.97 (0.89–1.06)0.46
Obstructive LUTS1.01 (0.91–1.12)0.85
Previous IH repair2.55 (0.99–6.36)0.054
History of hiatal hernia1.09 (0.21–5.69)0.91
Smoking1.68 (0.79–3.59)0.18
PSA level1.10 (0.97–1.23)0.13
Gleason score1.05 (0.50–2.20)0.90
Clinical stage1.24 (0.56–2.75)0.60
Prostate volume1.01 (0.99–1.03)0.49
Pathological stage0.90 (0.42–1.97)0.80)

The sensitivity and NPV of PE, USUS and MRI to diagnose a preoperative IH are presented in Table 3. USUS was the best procedure for preoperatively diagnosing IHs. The combination of MRI and USUS detected all but three IHs. The concordance between these three diagnostic methods was compared for pair of methods to give a percentage agreement and κ coefficient. USUS and PE exhibited moderate agreement (87.2%, κ 0.55) while MRI only exhibited fair agreement with both USUS and PE (77.4%, κ 0.28 and 80.5%, κ 0.20, respectively).

Table 3.  Detection of IH by PE, USUS, and MRI
Testn/N (%)
IHsSensitivityNPVIH detected by test alone
PE27/226 (12.0)27/72 (37.5)154/199 (77.4) 3/72 (1.4)
USUS50/226 (22.1)50/72 (69.4)154/176 (87.5)18/72 (25.0)
MRI37/226 (16.4)37/72 (51.4)154/189 (81.5)19/72 (26.4)
PE + USUS55/226 (24.3)55/72 (76.4)154/171 (90.0)
PE + MRI54/226 (23.9)54/72 (75.0)154/172 (89.5)
USUS + MRI69/226 (30.5)69/72 (95.8)154/157 (98.1)

DISCUSSION

We have previously reported that an appropriate inguinal PE will detect IHs in ≈13% of men undergoing ORRP [10]. This detection rate corresponds to the incidence of symptomatic IHs that develop after RP when no effort is made to detect these pre-existing IHs. The overwhelming majority of IHs detected preoperatively were asymptomatic and would not have been observed if an appropriate inguinal PE was not performed. All of these observations suggest that ORRP rarely causes IHs but more likely transforms an asymptomatic IH into a symptomatic IH.

The ability to perform simultaneous preperitoneal IHR at the same time as the RP [7,8] provides the justification for maximizing the detection of these hernias preoperatively. When preperitoneal herniorrhaphies are performed on those hernias diagnosed by an inguinal PE, the incidence of symptomatic post-prostatectomy hernias drops to 8.4%[9]. It appears that either PE does not maximally detect pre-existing IHs or in some cases the surgical procedure in a subset of cases causes an IH.

We have previously reported that routine abdominal CT lacks sensitivity for the preoperative detection of IHs [9]. The lack of sensitivity is most probably because the imaging study is performed supine without a Valsalva manoeuvre.

A study by van den Berg et al. [11] examined the diagnostic accuracy of PE, US, and MRI in 41 patients referred for suspected groin hernias. The ‘gold standard’ for diagnosis of IH was intraoperative examination of both groins at the time of laparoscopic hernia repair. Van den Berg et al. found PE to have the lowest sensitivity (74.5%) and the same specificity as MRI (96.3%). The sensitivity of US was similar to MRI (92.7% and 94.5%, respectively) but its specificity was inferior to MRI (81.5%). That study concluded that although MRI has a better positive predictive value and NPV, US should be the diagnostic method of choice given its low cost and availability [11].

Several more recent studies evaluating the accuracy of US for the diagnosis of IHs in patients with clinically suspicious groin hernias, later verified during surgical intervention, show US overall to be an accurate diagnostic method [12–15]. The sensitivity and specificity of these studies ranged from 33% to 100% and 87% to 100%, respectively [12–15].

Unlike these previous studies, the present series is the first to examine the ability to detect IHs in an unselected cohort of men undergoing ORRP using PE, USUS, and fast sequence MRI during a Valsalva manoeuvre. One important caveat is that the evaluation for IH with MRI was performed supine as it is not possible to use this technique in the up-standing position. The Valsalva inguinal MRI had a low sensitivity (51.4%; Table 3). In all, 19 IHs were found on Valsalva MRI when both PE and USUS failed to show an IH. It is not surprising that the study by van den Berg et al. [11] found MRI to be a more sensitive diagnostic tool than the current series as their population presented with a clinical history consistent with IH. The vast majority of men in the present series had no complaints of groin pain and/or bulge.

Advantages of inguinal US over MRI include the lower cost and that the procedure can be performed in real-time in the up-standing position with Valsalva in the urologist’s office. However, disadvantages of US include the high operator dependency and suboptimal view in overweight and obese patients. Inguinal US performed with Valsalva in the standing position was the most sensitive (69.4%) method for detecting IHs in the present study (Table 3). However, the definition of IH used in the present study biases the sensitivity and NPV in favour of the method that has the highest prevalence of positive results. USUS diagnosed the most IHs (22.1%) vs PE (12.0%) and MRI (16.4%; Table 3). In all, 19 IHs were detected by USUS when MRI and PE were negative and three IHs were detected by PE when USUS and MRI were negative. USUS correlated highly with PE.

Nielsen and Walsh [10] reported that 33% of men will have an IH diagnosed intraoperatively. An obvious limitation of intraoperative diagnosis is the subjectivity of the examination. Despite this very high detection rate, 5.3% of men developed a hernia postoperatively [10].

A major strength of the present study was that all the individuals responsible for diagnosing IHs on PE, MRI, and USUS were ‘blinded’ to the interpretations of the other diagnostic methods. In addition, the studies were interpreted prospectively and by a single examiner for the three diagnostic methods.

One limitation of the present study is the lack of 1 year follow-up data to determine if adding an imaging method, e.g. US decreases the likelihood of developing a symptomatic IH after RP. Another limitation is the definition of IH used in the present study. Defining a true IH as any diagnosed by any method biases sensitivity and NPV in favour of the method that has the highest prevalence of positive results. Additionally, we cannot calculate specificity or positive predictive value without an independent reference standard to diagnose IHs.

Conflicting evidence exist regarding the association of history of prior IHR [2–4,16,17], lower BMI [3,4,9], increasing age [3,4,9,16,17] and post-prostatectomy anastomotic stricture [2,3,17] and the risk of IH after RP. One study reported a significant relationship between postoperative wound-related problems and the development of IH after RRP [16]. No study has ever shown evidence linking prostate volume [3,4,16], duration of surgery [17] or smoking history [16] to the development of IH after RP.

Others studying the development of IH in the general population have found increasing age [6,7], history of prior IHR [2], probably prostatic hypertrophy based on LUTS [7], history of hiatal hernia [6,18], varicose veins [7], haemorrhoids [7] and physical effort [19] to be associated with a higher incidence of IH, while Black race [6] and obesity [6] were associated with a lower incidence. Tobacco smoking [6,19], constipation [6,7] and chronic cough [6,7] have not been associated with an increased risk of IH.

In the present series, we examined multiple factors that appeared to be reasonable potential predictors of the presence of a preoperative IH (Table 2). None of these factors significantly predicted the finding of an IH before ORRP, although a history of prior IHR approached statistical significance. In our previous series, increasing age and lower BMI were associated with preoperative detection of IHs [9].

In conclusion, inguinal USUS during a Valsalva manoeuvre was the most sensitive method for diagnosing pre-existing IHs in surgical candidates for ORRP. Overall, 12.0% of groins in the present study had an IH on inguinal PE alone. USUS detected an additional 18 IHs missed by PE and MRI detected an additional 27 IHs missed by PE. About one quarter of IHs would have been missed if USUS or MRI had not been performed. Based on the present findings, we recommend that all surgical candidates for ORRP undergo evaluation for IH by an inguinal PE and at least one imaging method. All men with a diagnosed preoperative IH should undergo a preperitoneal inguinal herniorrhaphy to avoid the need for a second surgery. The urologist is encouraged to become familiar with this approach to herniorrhaphy.

CONFLICT OF INTEREST

Herbert Lepor is co-owner of Medreviews and a Consultant for Watson, Steba Pharma and Aeterna Zentaris. He is also both an Investor and Consultant for Serenity and USHIFU.

Ancillary