Introduction
Inguinal hernias are the most common form of hernias where a defect of the abdominal wall leads to protrusion of the parietal peritoneum with or without abdominal contents at the level of the groin [
1]. The lifetime risk of developing inguinal hernia is estimated to be 27–43% in men and 3–6% in women with increasing incidence at advanced age and with higher body mass index [
1‐
3]. Inguinal hernias can be complicated by incarceration or strangulation [
4]. Surgical repair is the treatment of choice for symptomatic disease.
According to current guidelines, clinical assessment remains the mainstay for diagnosing inguinal hernia, and imaging is seldom warranted [
5,
6]. Nonetheless, imaging can assist in diagnosing clinically occult cases and for surgical planning [
7,
8]. Furthermore, differentiating hernia types and confirming bilateral hernias through clinical examination is challenging, but important for planning treatment [
9,
10].
Ultrasound (US) with dynamic maneuvers (e.g., Valsalva maneuver) is most often used for the assessment of inguinal hernia [
11‐
13]. Valsalva maneuver increases hernia conspicuity as the hernia sac protrudes more under increased abdominal pressure. Drawbacks of US are its operator-dependence and limitations related to patient size. Moreover, interpretation of US images for treatment planning purposes can be challenging for the surgeon [
14,
15].
Magnetic resonance imaging (MRI) is less examiner-dependent and images are easier to interpret for non-radiologists given the possibility of multiplanar views [
16]. However, MRI has some contraindications (claustrophobia, non-compatible devices) and may not be readily available at every institution.
The literature about the accuracy of computed tomography (CT) imaging for the diagnosis of inguinal hernia is scarce. One study evaluated CT under Valsalva maneuver (Valsalva-CT) for the diagnosis of abdominal wall hernias [
17]. However, only five inguinal hernias were included and there was no reference standard to verify findings. Others investigated the diagnostic performance of CT in prone positioning for the diagnosis of inguinal hernias and found higher accuracy of prone compared to supine CT (98.1 vs. 72.8%) [
18,
19]. However, prone positioning may not be feasible for all patients and may mask the presence of other concurrent abdominal wall hernias.
We are routinely using a dedicated protocol consisting of a non-contrast CT of the abdomen/pelvis in supine position acquired during a Valsalva maneuver in patients with suspected inguinal hernia to screen for additional unsuspected occult hernias and other abdominal wall hernias as these could influence treatment planning. The primary objective of this study was to evaluate the diagnostic performance of Valsalva-CT for the diagnosis and characterization of inguinal hernias. Secondary objectives were to assess interreader agreement and to investigate influencing factors on the diagnostic performance of the modality.
Discussion
In this study we evaluated the diagnostic performance and interreader agreement of Valsalva-CT for the detection and characterization of inguinal hernias. This study showed substantial interreader agreement (α = 0.723), high specificity (92.5–98.1%) and accuracy (81.1–91.5%) of Valsalva-CT. In contrast, sensitivity was only moderate, ranging from 68.2% to 85.8%, which was associated with missed smaller hernias.
A few studies have investigated the role of CT for the diagnosis of inguinal hernia [
8,
16‐
19,
23‐
25]. However, those studies either included a small number of patients [
17,
23], did not investigate the Valsalva maneuver [
16,
18,
19], did not have a dedicated read-out of CT images [
16], did not clearly describe their reference standard [
17], or focused on differentiation between types of groin hernias [
24,
25]. In a systematic review, Piga et al. [
12] analyzed the existing literature on the diagnostic performance of different imaging modalities for the diagnosis of inguinal hernia. Sensitivity and specificity of CT ranged from 57 to 100% and 83 to 100%, respectively. Only one study was included where Valsalva maneuver was performed in a subset of 8 patients [
26].
Jaffe et al. [
17] examined CT with and without Valsalva maneuver for identifying abdominal wall hernias and found increased conspicuity of hernias with Valsalva maneuver. Notably, most cases involved abdominal wall hernias other than inguinal hernias. Furthermore, the reference standard is not clear and diagnostic performance metrics were not reported in a consistent manner. We included all surgically proven inguinal hernias, regardless of size, even those incidentally diagnosed during surgery for other hernias. Hence, our cohort also included patients with small inguinal hernias. The threshold for calling a hernia in subtle cases may vary among individual readers, which could account for the lower and more inconsistent sensitivity rates observed. This hypothesis is supported by the observation that the size of the hernia neck and sac were notably smaller in cases overlooked by all three readers, relative to those identified by all three readers.
In our study, the two radiologists who were more experienced and subspecialized in abdominal imaging had a higher accurary than the reader with experience but without subspecialization. These findings are in line with a previous study from Miller et al. [
8] where a dedicated second reading of hernia scans by a radiologist led to a significant increase in accuracy compared to the baseline radiology reports (accuracy increased from 35 to 79%). Other studies have shown an increase in perceived report quality and decrease in interpretive discrepancies with subspecialized reporting [
27,
28]. Subspecialized reporting can yield significant clinical advantages, particularly in instances involving small and clinically occult hernias.
Accuracy for hernia type characterization was only moderate (66.5%–84.3%). Direct and indirect inguinal hernias were classified more accurately than combined hernias. While 80% of combined hernias were recognized as a hernia, only 24% were accurately identified as a combined type. Combined hernias tend to have one component that is more pronounced than the other, potentially leading to a lower degree of accuracy in their classification [
8]. Kamei et al. [
18] reported correct classification in 95.8%, but with possible selection bias as only surgery was used as a reference standard and hernia size was not reported. Their data also suggests a higher misclassification rate for combined hernias. Clinically, the type of inguinal hernia may not be as important for treatment planning as the detection of any hernia or the accurate identification of hernia contents, as this may inform treatment urgency, surgical approach, and complexity [
29‐
32]. In our study, the detection rate of inguinal hernias containing bowel or bladder was high, ranging from 94.3 to 95.2%. Only three cases (2.9%) were overlooked by all three readers.
The recurrence rates following inguinal hernia repair can be as high as 15%, and their diagnosis through clinical examination alone can prove challenging owing to scarring and fibrosis [
32,
33]. Detection rates for inguinal hernias were similar between surgery-naïve patients and those with history of inguinal hernia repair in our cohort, indicating that Valsalva-CT can be used as a diagnostic tool in patients with suspected recurrent hernia.
Our study has several limitations. First, it is a single-center retrospective study with data from a tertiary referral center and hence our findings may not be representative of other clinical practices and results should be validated in a prospective multi-center setting. We tried to minimize this bias by including all patients in a consecutive manner. Nevertheless, our study includes to date the largest cohort investigating supine CT with Valsalva maneuver for the detection of inguinal hernias. Second, surgery was performed by different surgeons from our hospital and varying surgical techniques may have introduced heterogeneity. However, as opposed to reoperation rates and outcome, which are established surgeon quality metrics, intraoperative confirmation of an inguinal hernia is less likely to be affected by surgeon-level variation [
34]. Furthermore, we used a composite clinical reference standard which entails the risk of over- or underestimation of the diagnostic test accuracy, can introduce verification bias, and affects the translation of our findings into a broader clinical setting. However, in the absence of a perfect reference standard we defined our reference standard based on current clinical practice guidelines for the diagnosis of inguinal hernia [
5,
20]. The hernia size and European Hernia Society (EHS) groin hernia classification [
5] was not consistently reported in the operative reports. Therefore, we were not able to correlate imaging-based hernia size with objective measurements from surgery in this retrospective study. Last, given the retrospective nature of the study, we were not able to compare the diagnostic accuracy of CT with other imaging modalities like US or MRI.
In conclusion, CT with Valsalva maneuver has a very high specificity and high accuracy for the diagnosis of inguinal hernia and the detection rate for inguinal hernias containing bowel or bladder is high. However, sensitivity is only moderate and affected by hernia size, as very small hernias are more frequently missed.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.