MDCT of Abdominal Wall Hernias: Is There a Role for Valsalva s Maneuver?

Transcription

1 bdominal Imaging Jaffe et al. s Maneuver in MDCT of Hernia Tracy. Jaffe 1 Martin J. O Connell John P. Harris Erik K. Paulson David M. DeLong Received May 10, 2004; accepted after revision July 23, ll authors: Department of Radiology, Duke University Medical Center, ox 3808, Durham, NC ddress correspondence to T.. Jaffe (jaffe002@mc.duke.edu). JR 2005;184: X/05/ merican Roentgen Ray Society MDCT of bdominal Wall Hernias: Is There a Role for s Maneuver? OJECTIVE. Our objective was to evaluate the role of s maneuver during MDCT for the diagnosis and characterization of abdominal wall hernias. SUJECTS ND METHODS. From September 2002 to May 2003, 100 consecutive patients (37 men and 63 women; mean age, 53 years) with suspected anterior abdominal wall hernias underwent 4-, 8-, or 16-MDCT with and without s maneuver. Patients received both oral and IV contrast material. On a workstation, three independent reviewers evaluated each scan obtained during rest and during s maneuver for the following parameters: anteroposterior (P) diameter of the abdomen; presence, location, and contents of the hernia; and transverse diameter of the fascial defect. The scans were compared to assess for changes in hernia size and contents and to determine whether the hernia would have been overlooked without s maneuver. Fisher s exact test, the McNemar test, and Cohen s kappa coefficient were used to assess for significant differences. RESULTS. The three reviewers identified a mean of 72 abdominal wall hernias (72%). The reviewers agreed (κ = 0.723) with respect to the presence of a hernia. P diameters increased an average of 1.33 cm during s maneuver (p < 0.001). The transverse diameter of the fascial defect increased an average of 0.66 cm and the P diameter of the hernia sac increased an average of 0.79 cm during s maneuver (p < 0.001). Fifty percent of the hernias became more apparent with s maneuver. Ten percent of the hernias could be detected only on the scan obtained during s maneuver. Conversely, in no patients was the hernia detected only on the rest scan. CONCLUSION. s opposed to scans obtained at rest, scans obtained during s maneuver aid in the detection and characterization of suspected abdominal wall hernias. single scan obtained during s maneuver is sufficient to detect 100% of anterior abdominal wall hernias identified on CT. he repair of abdominal wall hernias has become a growth area for T many surgical practices [1, 2]. Implicit in the workup of these hernias is the ability to diagnose and characterize the size and specific location of the anatomic defect and the contents of the hernia sac. The diagnosis of abdominal wall hernias is based primarily on physical examination, which includes s maneuver. s maneuver is used to make the hernia more apparent clinically, because the increase in intraabdominal pressure causes the hernia sac to enlarge and protrude through the anatomic defect [3]. The physical examination is made more difficult in obese or postoperative patients, in whom it is difficult to distinguish a hernia from a postoperative complication such as a hematoma, abscess, or seroma [4 19]. Patients are often referred for imaging studies to confirm a suspected hernia. These studies include radiography, sonography, barium studies, and herniography. CT is being used increasingly to show the specific anatomy of the hernia sac and distinguish a hernia from its mimics. lthough s maneuver is often used in barium studies or sonography of the abdominal wall, we found no research that evaluated the role of s maneuver in MDCT diagnosis of abdominal wall hernias [1, 2]. With MDCT, the short acquisition time allows imaging of the entire abdomen and pelvis within a single breath-hold including s maneuver. The aim of our study was to determine the effect and benefit of s maneuver in the MDCT diagnosis of abdominal wall hernias. JR:184, March

2 Jaffe et al. Subjects and Methods This study was approved by the institutional review board. From September 2002 to May 2003, 100 consecutive patients with suspected abdominal wall hernias underwent MDCT (37 men and 63 women; age range, years; mean, 53 years). Scanning was performed from the dome of the diaphragm to the inferior margin of the pubic symphysis with a 4-, 8-, or 16-MDCT scanner. Patients ingested 450 ml of a 2% barium sulfate suspension (Readi-Cat Z, E-Z-EM) or dilute meglumine diatrizoate (Gastrografin, ristol-myers Squibb) 1 2 hr before imaging. The initial rest acquisition was performed at end-inspiration during a single breath-hold. For this acquisition, 150 ml of iopamidol (Isovue 300, ristol-myers Squib) was injected at a rate of 3 ml per second, unless contraindicated. Seventy-three of the 100 patients received IV contrast material. Scans were acquired during the portal venous phase of enhancement, with a delay determined by bolus tracking and automated triggering technology. Protocols varied depending on whether a 4-, 8-, or 16-MDCT scanner was used. Pitch ranged from 1.5 to 1.75, and table speed ranged from 15 mm per rotation to 17.5 mm per rotation. ll scanners had a 0.5-sec tube rotation. Slice reconstruction thickness was 5 mm for all patients. cquisition duration ranged from 12 to 20 sec. Immediately after the initial scanning, at end-inspiration, the patient was instructed to bear down or strain and a second acquisition was completed during s maneuver. No additional IV or oral contrast material was administered for the second acquisition. The technical parameters for the scan obtained during s maneuver were identical to those of the scan obtained at end-inspiration. Three experienced radiologists with subspecialty training in abdominal imaging reviewed, on a workstation (Centricity 1.0, GE Healthcare), each scan obtained at rest and during s maneuver. The reviewers were unaware of patient identifiers and interpreted each scan independently. Clinical information was not provided to the reviewers. Most notably, reviewers were unaware of the patients prior surgical histories. Each scan was evaluated for multiple parameters including presence, location, and contents of the hernia; anteroposterior (P) diameter of the abdomen; and transverse diameter of the fascial defect. The P diameter of the abdomen was determined by measuring the distance from the anterior aspect of the L5 vertebral body to the anterior skin surface. This distance was measured on both scans. For the purposes of this study, a hernia was defined as a protrusion of abdominal contents through a defect in the normally restraining muscles of the abdomen and pelvis. The location of the hernia was designated as ventral, inguinal, or other. Contents were designated mesenteric or omental fat, colon, small bowel, stomach, bladder, or other. The transverse diameter of the hernia sac aperture and the P diameter of the hernia sac were measured on both scans. The scans obtained at rest and during s maneuver were compared to assess changes in hernia size and contents and to determine whether the hernia was more conspicuous after s maneuver. The reviewers determined whether the hernia would have been overlooked without s maneuver. In addition, they assessed whether evidence of a prior hernia repair was present or whether postoperative complications were present. In the patient with multiple hernias, the reviewers were instructed to characterize the largest defect. The data were collected and entered into a database by the principal investigator. No clinical information was included in this study. The department statistician analyzed the data using version 8.2 of the SS software system (SS Institute). The statistical reliability of a given difference in response rates between respondent subgroups for specific questions was analyzed with Fisher s exact test, the McNemar test, and Cohen s kappa coefficient. p value of less than 0.05 was considered statistically significant. kappa value of greater than 0.5 supported agreement between reviewers. Results The three reviewers, designated as R1, R2, and R3, identified an abdominal wall hernia in 69, 74, and 72, respectively, of the 100 patients (mean, 72 hernias). The reviewers agreed with respect to the presence of a hernia (κ = 0.72). Most hernias were ventral (Fig. 1) (R1, 63 [91%]; R2, 66 [89%]; R3, 67 [93%]); Fig year-old man with suspected ventral hernia after abdominal surgery., CT scan obtained with IV and oral contrast material at rest shows defect in anterior abdominal wall (arrows). No intraperitoneal contents herniate through defect. Second defect in muscular anterior abdominal wall is seen (arrowheads) just craniad to ileostomy. t rest, no intraperitoneal contents protrude through this defect., CT scan obtained during s maneuver shows protrusion of colon through midline defect (arrows). In addition, small bowel protrudes (arrowheads) at peristomal defect. 848 JR:184, March 2005

3 s Maneuver in MDCT of Hernia the inguinal hernias were fewer (R1, 5 [7%]; R2, 7 [9%]; R3, 4 [5%]). One hernia was posterior intercostal. The reviewers did not differ in the identification of hernias between men and women. s maneuver was associated with an increase in P diameter in 82 patients (82%) (Fig. 2), a decrease in 14 patients (14%), and no change in four patients (4%). The mean increase in diameter was 1.40 cm, as detailed in Table 1. The change in mean P diameter was statistically significant for all three reviewers (p < 0.001). The P diameter of the hernia sac increased by an average of 0.8 cm with s maneuver (p < 0.001). With s maneuver, the transverse diameter of the hernia sac increased in 51 patients (72%) (Fig. 2), decreased in 11 patients (15%), and did not change in 10 patients (14%). For all three reviewers, the mean change in transverse diameter was 1.40 cm, which was statistically significant (p < 0.001). Hernias became more conspicuous after s maneuver (Table 2). Reviewers noted that the appearance of 56 77% (mean, 70%) of the hernias changed with s maneuver, and 32 72% (mean, 36%) of the hernias became more apparent with s maneuver. The contents of the hernia sac changed with s maneuver, as the maneuver caused small bowel, colon, liver, or fat to protrude into the hernia (Fig. 3). Most important, R1, R2, and R3 found that the hernias in 10, seven, and four patients, respectively, (mean, 7) would have been overlooked if s maneuver had not been not performed (Fig. 4). Conversely, in no patient was a hernia detected only on a rest scan. R1, R2, and R3 identified 32 of 68, 37 of 74, and 46 of 72 hernias, respectively, as craniad to the umbilicus. The remaining hernias were caudad to the umbilicus. For each reviewer, the effect of s maneuver, change in P diameter, change in transverse measurement of the fascial defect, and conspicuity were similar regardless of whether TLE 1 Reviewer Fig year-old man with suspected abdominal wall hernia., CT scan obtained with IV and oral contrast material at rest shows deficiency of anterior abdominal wall (arrows) with protrusion of small bowel and omental fat. Horizontal line denotes transverse diameter of hernia aperture. Short vertical line denotes anteroposterior (P) measurement of hernia sac; long vertical line denotes P measurement of abdomen., CT scan obtained during s maneuver shows protrusion of mesenteric fat and additional small bowel into hernia sac (arrows). In addition, transverse diameter of hernia aperture is wider (horizontal line). Note increase in P measurement of hernia sac (short vertical line) and P diameter of abdomen (long vertical line) with s maneuver. the hernia was craniad or caudad to the umbilicus (p > 0.2 in all cases). Similarly, whether the hernia was craniad or caudad to the umbilicus had no bearing on the likelihood that a hernia would be missed without the s maneuver scan (p > 0.4). In 24% of the patients scanned, prior hernia repair was evident through the presence of mesh or coils on CT. The reviewers identified 14 postoperative complications, including seroma, hematoma, and abscess. Discussion The identification and repair of abdominal wall hernias have become an area of great interest and growth in surgical programs Effect of s Maneuver on bdominal Diameter and Transverse Diameter of Hernia Sac efore P Diameter fter Change p efore Transverse Diameter fter Change R < < R < < R < Mean N N Note. Values are in centimeters. N = not applicable. p JR:184, March

4 Jaffe et al. Fig year-old man with suspected ventral hernia., CT scan obtained at rest with IV and oral contrast material shows defect (arrows) in upper abdominal wall. Portion of colon protrudes through defect, CT scan obtained during s maneuver shows no change in appearance of hernia (arrows). throughout the world [1, 2, 4 17]. Most abdominal wall hernias are obvious clinically. However, the physical examination may be unreliable in obese patients and patients who have had a prior herniorraphy. Studies have found cross-sectional imaging to be sensitive and specific for the evaluation and characterization of hernias, especially in obese or postoperative patients, in whom the clinical diagnosis of recurrence may be difficult [4 19]. MDCT acquisition times have decreased to such a degree that it is possible to scan the entire abdomen and pelvis during a single s maneuver. lthough the addition of s maneuver during CT has been suggested, to our knowledge, no study has documented the efficacy of this maneuver in the diagnosis of abdominal wall hernia. ll our patients received oral and IV contrast material unless clinically contraindicated. Oral contrast material helps distend bowel loops and enhances the differentiation of small bowel from colon. We used IV contrast material to help distinguish between the presence of Fig year-old woman after mesh placement for hernia repair., CT scan obtained at rest with IV and oral contrast material shows mesh (arrows) in anterior abdominal wall. No recurrent hernia is evident at rest., CT scan obtained during s maneuver shows herniation of portion of gas-filled small bowel (arrowheads) at left aspect of mesh (arrow). In this patient, recurrent hernia would have been overlooked unless s maneuver scan had been obtained. 850 JR:184, March 2005

5 s Maneuver in MDCT of Hernia a hernia and its mimics, to assess for inflammation, and to increase the sensitivity for incidental abnormality in the solid organs [11, 16]. s maneuver was originally described in 1704 by ntonio Maria, an Italian anatomist who developed this method for inflating the middle ear. y creating force against a closed glottis, this exercise increases pressure within the thoracic and abdominal cavity. The proposed and desired effect of s maneuver in this study was to increase intraabdominal pressure and accentuate any herniation that may have been occult at rest [3]. Ventral hernias have been evaluated using both rest scans and s maneuver scans at our institution for a number of years. This study was designed to evaluate the role of a scan obtained during s maneuver. Our study showed that s maneuver has an effect on the abdomen and its contents that aids in the detection and characterization of anterior abdominal wall hernias. To our knowledge, no published norms exist on the expected change in intraabdominal morphology on CT after s maneuver. We hypothesized that measurement of P diameter changes would reflect an increase in intraabdominal pressure. In patients with or without identifiable hernias, P diameter increased during s maneuver, suggesting an increase in intraabdominal pressure and producing a satisfactory result from the s maneuver. This was a consistent observation and provided evidence that patients can be instructed in the performance of s maneuver and can maintain the maneuver throughout an MDCT scan. With s maneuver, the transverse measurement of the fascial defect increased. The change in size implies that the abdominal wall musculature changes with s maneuver. These changes accentuate the aperture of the hernia sac. In addition, the P diameter of the hernia sac changes with the addition of s maneuver, implying that the increase in intraabdominal pressure also is transmitted to the hernia sac and its contents. More important, a mean of 51% of hernias become more conspicuous with s maneuver because of protrusion of mesenteric or omental fat, bowel, or solid organs into the hernia sac. s maneuver simulates certain activities of daily living such as heavy lifting, making these findings more clinically relevant. Most important, there are hernias (10% in this series) that will not be diagnosed unless a scan is obtained with s maneuver. These defects are subtle. t rest, the hernia sac is confined within the abdominal wall and not TLE 2 Reviewer Effect of s Maneuver on Hernia Conspicuity Number of Hernias Change with s Maneuver detected. It is only with the increased pressure from s maneuver that these fascial defects and hernia sacs become identifiable. The identification of this subset of patients provides a strong argument that in patients referred with suspected hernia, it is imperative to obtain a scan during s maneuver. bsence of hernia on a scan obtained at rest cannot exclude the presence of a hernia. This study had some limitations. First, the s maneuver scan was obtained immediately after the rest scan. The scans were not identical because no additional IV contrast material was administered for the second scan. We do not think that this discrepancy in technique affected the final characterization of the anterior abdominal wall. Second, the rest scan and the s maneuver scan were interpreted at the same time. This design may have created a bias that would not be present if the studies were interpreted individually and randomly. Third, we reviewed only 100 patients, and only ventral hernias were well represented. Finally, we did not study the clinical impact of the noted increase in hernia conspicuity. In our series, no patient showed a hernia on the rest scan but no hernia on the s maneuver scan. This finding suggests that a single scan during s maneuver would identify most, if not all, anterior abdominal wall hernias. On the basis of our results, we have altered our practice: In patients referred for suspected hernia, we obtain a scan only during s maneuver, as opposed to both at rest and during s maneuver. This approach will reduce the radiation dose to the patient and lower the cost, scanner time, and archive memory. single scan obtained during s maneuver is sufficient to detect 100% of anterior abdominal wall hernias identified on CT. More Conspicuous with s Maneuver Identified Only with s Maneuver R (75) 49 (71) 10 (14) R (77) 37 (50) 7 (9) R (56) 23 (32) 4 (6) Mean (69) 36 (50) 7 (10) Note. Numbers in parentheses are percentages. References 1. Gore RM, Levine MS, eds. Textbook of gastrointestinal radiology, 2nd ed. Philadelphia, P: Saunders, 2000: Schwartz SI, Shires GT, eds. Principles of surgery, 6th ed. New York, NY: McGraw-Hill Inc., 1994: Light HG, Routeledge J. Intra-abdominal pressure factor in hernia disease. rch Surg 1965;90: Lee GHM, Cohen J. CT imaging of abdominal hernias. JR 1993;161: Toms P, Cash CCJ, Fernando, Freeman H. bdominal wall hernias: a cross-sectional pictoral review. Semin Ultrasound CT MR 2002;23: Wechsler RJ, Kurtz, Needleman L, et al. Cross-sectional imaging of abdominal wall hernias. JR 1989;153: Hojer M, Rygaard H, Jess P. CT in the diagnosis of abdominal wall hernias: a preliminary study. Eur Radiol 1997;7: Stable Ianora, Midiri M, Vinci R, Rotondo, ngelelli G. bdominal wall hernias: imaging with spiral CT. Eur Radiol 2000;10: Rose M, Eliakim R, ar-ziv Y, Vromen, Rachmilewitz D. bdominal wall hernias: the value of computed tomography diagnosis in the obese patient. J Clin Gastroenterol 1994;19: Gutierrez del la Pena G, Vargas Romero J, Dieguez Garcia J. The value of CT diagnosis of hernia recurrence after prosthetic repair of ventral incisional hernias. Eur Radiol 2001;11: Lin HJ, Vargish T, Dachman H. CT findings after laparoscopic repair of ventral hernia. JR 1999;172: Emby DJ, oun G. CT technique for suspected anterior abdominal wall hernia. JR 2003;181: Toms P, Dixon K, Murphy JMP, Jamieson NV. Illustrated review of new imaging techniques in the diagnosis of abdominal wall hernias. r J Surg 1999;86: Harrison L, Keesling C, Martin NL, Lee KR, Wetzel LH. bdominal wall hernias: review of herniography and correlation with cross-sectional imaging. RadioGraphics 1995;15: Miller P, Mezwa DG, Feczko PJ, Jafri ZH, Madrazo L. Imaging of abdominal hernias. RadioGraphics 1995;15: Goodman P, alachandran S. CT evaluation of the abdominal wall. Crit Rev Diagn Imaging 1995;33: Ghahremani GG, Jimenez M, Rosenfeld M, Rochester D. CT diagnosis of occult incisional hernias. JR 1987;148: Stamm ER, Pretorius DH, Olson LK. bdominal wall CT: a pictorial essay. Comput Radiol 1985;9: Zarvan NP, Lee FT, Yandow DR, Unger JS. bdominal hernias: CT findings. JR 1995;164: JR:184, March