Materials and Methods

Inclusion criteria—Medical records of dogs with perineal hernia surgically treated by IOT at Colorado State University Veterinary Teaching Hospital from August 30, 1998, to August 7, 2012, were reviewed. Signalment, including age, sex and neuter status, and breed, was recorded for each patient. Clinical signs and diagnostic methods used to reach a definitive diagnosis were recorded. Dogs with < 15 days of follow-up were excluded from the study.

Preoperative scoring—Dogs were assigned a preoperative clinical sign score on the basis of the duration and severity of signs associated with the gastrointestinal tract and the lower urinary tract (Appendix 1). A scoring method described by Orsher¹⁶ was used. Each problem of the gastrointestinal tract (diarrhea, tenesmus, dyschezia, fecal incontinence, fecal impaction, and hematochezia) and the lower urinary tract (dysuria, stranguria, hematuria, pollakiuria, and urinary incontinence) were assigned a numerical value: 0, problem absent; 1, occurrence up to 7 times per week with small volume (where applicable), little signs of pain or discomfort, and minor seriousness; 2, occurrence up to 7 times per week with larger volume (where applicable), more signs of pain or discomfort, and greater seriousness; and 3, occurrence with each defecation or urination. The sums of individual clinical problem scores were calculated to create a composite preoperative clinical sign score for each patient.

Postoperative scoring—Postoperative information for patients was obtained from follow-up examinations at Colorado State University or with the referring veterinarian or by telephone interview of the owners (Appendix 2). A postoperative clinical sign score was assigned to patients also following the method of scoring described by Orsher.¹⁶ Numerical values were assigned to each patient on the basis of the duration of gastrointestinal tract and lower urinary tract problems following surgery. Scores were assigned as follows: 0, problems occurring up to 4 weeks after surgery; 1, > 4 to 8 weeks after surgery; 2, > 8 to 16 weeks after surgery; and 3, > 16 weeks after surgery. The sums of individual problems scores were used to create a postoperative clinical sign score for each patient.

Postoperative complications strictly related to perineal herniation and repair included rectal prolapse, tenesmus, stranguria, hematochezia, dyschezia, fecal impaction, diarrhea, urinary incontinence, fecal incontinence, urinary tract infection, megacolon, and recurrence. Other complications like seroma, dehiscence, and infection were not considered specific to the repair of the hernia. Recurrence was defined as the development of perineal hernia on the ipsilateral side any time after correction of the initial hernia. Perineal hernia in the contralateral side after correction of the initial hernia was not considered as a complication of the initial surgery.

Statistical analysis—Descriptive statistics were calculated for different patient variables. A Fisher exact test was used to compare frequency distributions. Actuarial Kaplan-Meier analysis was used to identify the 1-year complication rate and recurrence rate. A univariate Cox analysis was used to identify risk factors associated with complications and recurrence. Results were considered significant at a value of P < 0.05.

Results

Review of the medical records revealed 62 dogs that had perineal hernia. Twenty-eight of the dogs did not have follow-up for at least 15 days after surgery; therefore, 34 dogs met the inclusion criteria to be entered in the study. Internal obturator muscle transposition was used in each case as a primary repair. Cystopexy was performed in 1 dog in conjunction with bilateral perineal hernia repair at the time of initial surgery. Twenty of the 34 IOT procedures were performed by a faculty member; the remaining 17 procedures were performed by surgical residents under close supervision of a faculty member. Preoperative clinical score was not significantly (P = 0.987) different between dogs that underwent surgery performed by faculty (median, 6; range, 0 to 14) and dogs that underwent surgery performed by residents (median, 6; range, 0 to 11). Median follow-up time was 345 days (range, 22 to 1,423 days), and mean ± SD follow-up time was 417 ± 401 days. Fifteen dogs were lost to follow-up by 529.2 ± 116.9 days. Three dogs were euthanized for reasons related to the perineal hernia. Two dogs were euthanized for recurrence and 1 dog for urinary incontinence. Fourteen dogs were still alive at the time of the study. Twenty-two dogs met the minimum 16-week follow-up standard set for postoperative clinical sign scoring.¹⁶ Referring veterinarians of 21 dogs were contacted. Owners of 16 dogs were also contacted.

Median age was 8.5 years (range, 6 to 13 years). All patients were male, with 22 of 34 (64.7%) sexually intact at the time of diagnosis. Of the 12 patients that were castrated, 5 (41.7%) had been castrated < 1 year before a diagnosis of perineal hernia was made. Affected dogs included 4 mixed-breed dogs, 3 Shetland Sheepdogs, 3 Maltese, 2 Collies, 2 Boston Terriers, 2 Australian Shepherds, 2 Border Collies, 2 Labrador Retrievers, and 1 of each of the following breeds: Beagle, Miniature Poodle, Pekingese, Husky, Pug, Miniature Dachshund, Toy Poodle, Brittany Spaniel, Doberman Pinscher, Rottweiler, West Highland White Terrier, and Scottish Deerhound. Breed was not recorded for 2 dogs.

Clinical signs included straining to defecate (n = 7), perineal swelling (6), straining or inability to urinate (2), and rectal prolapse (2). All dogs underwent a rectal examination, and 11 dogs had a diagnosis of perineal hernia made on the basis of rectal examination findings alone. Thirteen dogs had bilateral perineal hernias. Twenty-one dogs were had unilateral herniation, of which 16 (76%) were right sided. Additional diagnostic methods included abdominal radiography (n = 19), contrast radiography (ie, cystourethrogram, urethrogram, and colonogram; 4), and ultrasonography (9). Clinical preoperative signs included fecal impaction, tenesmus, hematochezia, dyschezia, diarrhea, hematuria, stranguria, urinary obstruction, and urinary incontinence (Table 1). The most common clinical sign was tenesmus (28/34 [82.4%]). Median preoperative clinical sign score was 6 out of a possible 33 (range, 0 to 14). Bladder retroflexion was significantly (P = 0.008) associated with a higher rate of urinary obstruction. Periprostatic fat and omentum (n = 34), urinary bladder (9), colon (6), prostate (4), and small intestines (1) were found in the hernia at the time of surgery.

Overall complications occurred in 14 dogs. Surgical complications not specific to perineal hernia were seroma (n = 3), infection (2), and incisional dehiscence (1). Postoperative complications related to the perineal hernia occurred in 10 of 34 dogs. This included tenesmus (n = 9), dyschezia (7), fecal impaction (3), stranguria (4), hematochezia (2), urinary incontinence (2), diarrhea (1), urinary tract infection (1), and megacolon (1). None of the dogs had fecal incontinence after surgery. Median time to complication was 52 days (range, 1 to 807 days). One dog that developed urinary incontinence had its bladder retroflexed in the hernia at the time of surgery. Median postoperative clinical sign score was 0 (range, 0 to 12) for the 22 dogs with > 16 weeks of follow-up. After surgery, the most common persistent signs were tenesmus and dyschezia. No risk factors were identified to be in association with the occurrence of complications (tenesmus, hematochezia, dyschezia, fecal impaction, rectal prolapse, urinary incontinence, or fecal incontinence; Table 2). One year after surgery, 51.2% of dogs were free of complication (Figure 1).

View Full Size

Kaplan-Meier curve of the proportion of dogs free of complications over time following surgical correction of perineal hernia by IOT. At 365 days after surgery, 51.2% dogs were free of complications. Diamonds represent each dog.

Citation: Journal of the American Veterinary Medical Association 246, 3; 10.2460/javma.246.3.321

• Download Figure
• Download figure as PowerPoint slide

Recurrence occurred in 7 of 34 (20.5%) dogs with a 1-year recurrence rate at 27.4% (Figure 2). Six of those dogs were castrated at the time of the first surgery. The other dog was castrated 1.5 months prior. Median time for recurrence to develop was 28 days (range, 2 to 364 days). Six of the dogs that had recurrence had surgeries performed by a faculty member; 2 had been performed by surgical residents under supervision by a faculty member. Recurrences were treated with cystopexy and colopexy in 2 dogs and mesh implantation in 1 dog. The remaining 5 dogs with recurrence did not undergo additional surgery. The mesh repair failed 2 weeks after surgery, and a primary herniorrhaphy was successfully performed then. Perineal herniation occurring opposite to the surgical site was successfully repaired with IOT in 2 dogs. Presence of postoperative tenesmus was identified as a significant (P = 0.037) risk factor for the development of recurrence (hazard ratio, 2.29; Table 3). Postoperative tenesmus was observed significantly (P = 0.049) more frequently in dogs that had a recurrence (4/7) than in dogs that did not (5/27). Bladder retroflexion at the time of initial evaluation or surgery could not be evaluated as a risk factor for the development of recurrence because none of those dogs had a recurrence. Three dogs developed a perineal hernia in the contralateral side between 35 and 95 days after surgery.

View Full Size

Kaplan-Meier curve of the proportion of dogs free of recurrence over time following surgical correction of perineal hernia by IOT. The 1-year (365-day) recurrence rate was 27.4%. Diamonds represent each dog.

Citation: Journal of the American Veterinary Medical Association 246, 3; 10.2460/javma.246.3.321

• Download Figure
• Download figure as PowerPoint slide

Discussion

The most common complications following IOT in this study were tenesmus, dyschezia, and recurrence of perineal hernia. Findings in this study indicate that recurrence can develop within at least 1 year after surgery. No significant risk factors were identified to contribute to the development of complications. However, presence of postoperative tenesmus was associated with an increased risk of recurrence.

The age, sex, and neuter status of affected dogs in this study were consistent with previously examined populations affected by perineal hernia.^{2,5,8,10,17,19,26} Older sexually intact male dogs were the only dogs in our study (median age, 8 years). The most commonly affected breeds observed were Shetland Sheepdog, Maltese, and mixed breed. Collies have been identified as a commonly affected breed.^2,8,17 Other breeds noted to be overrepresented include Corgi, Australian Kelpie, Poodle, Dachshund, Old English Sheepdog, Pekingese, and Boston Terrier.^2,5,8,10,17 A predilection of right-sided over left-sided hernias was identified in the present study as well as in others.^2,5,8,19

In the present study, clinical signs following IOT were improved and the postoperative clinical sign score was reduced, compared with the preoperative clinical sign score. The overall complication rate (14/34 [41.1%] dogs) and the rate for complications related to perineal hernia (10/34 [29.4%] dogs) were within the range of complication rates reported in the literature.^{2,5,10,12,13,17,19,21,23,25,26} The most common complications identified were persistent tenesmus and dyschezia after surgery. Tenesmus and incontinence (fecal or urinary) have been the most common complications reported in literature.^{2,10,19,23–26} In our study, 27% of the dogs had some tenesmus after surgery, which is within the range of the rates reported.^{2,10,16,19,24–26} Brissot et al²⁵ reported postoperative tenesmus in 44% of dogs after colopexy and cystopexy. Given that all the studies reported in the literature have been retrospective, this disparity in rate of tenesmus can be the result of the evaluation of tenesmus by owners. No significant risk factors contributing to the development of complications were identified in the study presented here. One year after surgery, 51.2% of dogs were free of clinical signs related to perineal hernia. Grand et al²⁶ reported that 50% of their case animals were free of clinical signs related to perineal hernia 80 months after surgery. Six dogs in their study developed complications between 30 and 80 months after surgery. In the present study, because dogs lost to follow-up were censored for the complication-free rate, the sample complication rate is a more accurate representation of the true complication rate after perineal hernia repair in dogs. Most of the complications developed within the first 3 months after surgery. Interestingly, 3 dogs still had complications or persistence of clinical signs related to the perineal hernia > 6 months after surgery; complications included fecal impaction, hematochezia, and dyschezia. Fecal impaction, tenesmus, hematochezia, and dyschezia are clinical signs related to colitis and proctitis, suggesting that dogs with perineal hernia should be evaluated for inflammatory bowel disease.^27,28 However, these clinical signs could also be related to rectal diverticula. To further investigate these clinical signs, a colonoscopy or a barium swallow test could be considered to determine whether rectal diverticula exist. Tenesmus could also result from a suboptimal hernia repair.

The 1-year recurrence rate was 27.4%. Rates of recurrence reported in the literature range from 0% to 46%.^{2,5,10,12–19,21,23,25} Those rates do not take into consideration the dogs lost to follow-up, which the 1-year recurrence rate reported in this study does. The 1-year recurrence rate in the present study is a better representation of the true recurrence rate after surgery than the standard recurrence rate because it adjusts the rate for dogs lost to follow-up. Obviously, it is not known whether dogs lost to follow-up developed recurrence. Cystopexy and colopexy have been used for treatment of perineal hernia in addition to IOT to reduce the risk of recurrence. Brissot et al²⁵ reported, in a group of 41 dogs, a recurrence rate of 10% after vasopexy, cystopexy, or colopexy and IOT, which is lower than the recurrence rate reported in the present study. Cystopexy and colopexy have been recommended to prevent herniation of the bladder and decrease pressure of the colon on the repair of the pelvic diaphragm.^22,25 Grand et al²⁶ did not show any benefit of colopexy and cystopexy for the rate of complications or recurrence after surgery even in dogs with bladder retroflexion. Case animals that underwent an initial abdominal surgery had more complications or recurrence in their study. Therefore, it is conceivable that abdominal surgery prior to IOT is not required even for complicated hernias as described by Brissot et al²⁵ However, additional comparative prospective studies would be beneficial to investigate this concept. Recurrence occurred > 6 months after the initial surgery, which illustrated the need for long-term follow-up to evaluate the outcome of surgeries for the treatment of perineal hernia.

Postoperative tenesmus was a negative prognostic indicator for the development of recurrence. Postoperative tenesmus more likely applies more pressure on the surgical repair, which would explain the increased risk of recurrence associated with tenesmus. Severity of tenesmus could not be evaluated because of the retrospective nature of the study. Also, postoperative tenesmus, dyschezia, and fecal impaction were more frequent in dogs that developed a hernia in the opposite side within 95 days after the first surgery. Therefore, it seems important to prevent postoperative tenesmus with appropriate diet and stool softener, as has been recommended.^27,28 It is also important to treat any inflammation and colitis that would also contribute to fecal impaction, hematochezia, and tenesmus.^29,30 Owing to the retrospective nature of the study, we could not evaluate whether tenesmus occurred during or after defecation. If tenesmus happens after defecation, it is more related to inflammation.³⁰

Preoperative clinical sign score and presence of a bilateral perineal hernia have been shown to be predictive of surgery outcome.^16,17 Hardie et al⁹ did not identify the presence of bilateral hernia as factors influencing outcome of the surgery. None of those variables appeared to be risk factor in the present study.

Surgeon experience has previously been suggested to play a role in the failure of the IOT repair.^2,5 However, lower surgeon experience level was not found to be a significant risk factor in the present study. The less experienced surgeons in our study were surgery residents who worked under close supervision during the procedure. This makes the difference in the level of experience between groups very limited and interferes with the effect experience level has on recurrence rate. The preoperative clinical score was not different for dogs undergoing surgery by faculty and the residents. However, that clinical score mostly addresses the severity of clinical signs and not necessarily the difficulty of the surgery. Hardie et al⁹ did not identify surgeon experience level as a factor influencing outcome of the surgery.

Bladder retroflexion at the time of surgery was not associated with an increased risk of recurrence; none of the dogs with bladder herniation had recurrence in this study. Perhaps, dogs with urinary bladder retroflexion require emergency surgical intervention. Therefore, they undergo surgical correction sooner than other dogs, preventing the development of chronic anatomic changes like rectal deviation that might place more pressure on the surgical site. No obvious correlation seems to be present between bladder retroflexion and recurrence rate, as reported in several studies.^4,12,19,24 In a study²⁶ on 41 cases of perineal hernia, bladder retroflexion was not associated with a higher rate of complications or recurrence.

Even with > 60% of cases in the literature involving a unilateral hernia, muscles in both sides of the pelvic diaphragms are atrophied and have signs of weakness resulting from the underlying cause of the perineal hernia.^{1,17,27,28,31} Three dogs developed a hernia in the contralateral side 1 to 3 months after the initial surgery. Because it happened at least 1 month after the first surgery, we did not consider herniation of the contralateral side as complications from the first surgery. Rectal examinations were performed at the time of initial evaluation and after induction of anesthesia, so the bilateral hernia was not present at the time of the first surgery. If the diagnosis of bilateral hernia had been missed, herniation would have occurred shortly after the first surgery. Also, dogs that developed a hernia in the contralateral side had a higher frequency of tenesmus, dyschezia, and fecal impaction.

This retrospective study evaluated only 1 surgical technique to treat perineal hernia. To make a better recommendation on the choice of surgical technique, it would have been valuable to have a group undergoing colopexy and cystopexy; however, this is rarely used in our institution. A prospective randomized clinical trial would be appropriate to compare the outcome of different surgical techniques. This was a retrospective study with inherent limitations. Fifteen dogs were lost to follow-up; however, their mean ± SD follow-up time was 529.2 ± 116.9 days, which was longer than the median follow-up time for the entire population. Actuarial Kaplan-Meier analysis with univariate Cox analysis was used to ameliorate this problem. Actuarial Kaplan-Meier survival curves were used to provide a description of the distribution of complications and recurrence over time. Follow-up evaluation was performed by different veterinarians and by owners, which may represent a bias in the level of complications reported. The severity of the complications could not be evaluated objectively because of the retrospective nature of the study. Medical records with review of operative report, information provided to the owner at the time of discharge from the hospital, and letter to the referring veterinarians were reviewed to increase the accuracy of the data entered in the study. The clinical sign score was used to compensate for the lack of information on the severity of the clinical signs. However, the presence or absence of clinical signs could still be evaluated to determine risk factors for recurrence or complications.

In conclusion, IOT for the surgical treatment of perineal herniation in dogs was associated with a 1-year recurrence rate of 27.4%. Clinical signs related to the perineal hernia can persist for > 1 year after surgery. Tenesmus after surgery was a risk factor for recurrence after treatment of perineal hernia with IOT.