Of the 17 dogs in which fistulas had completely healed, 9 did not have any evidence of recurrence between 15 and 19 months after cyclosporine administration was discontinued. One dog developed a new fistula 16 weeks after cyclosporine administration was discontinued. The fistula was at a site different from that of the original fistulas. The dog was treated with cyclosporine for an additional 13 weeks, and the fistula healed. The dog did not have any signs of recurrence more than 1 year after the end of the second course of cyclosporine treatment.
Fistulas recurred in 7 other dogs in which fistulas were completely healed after 16 weeks of cyclosporine treatment. Recurrences were detected between 2 and 24 weeks after cyclosporine administration was discontinued. One of the 7 dogs had fecal incontinence as a result of previous surgical treatment, and when the fistulas recurred, the owner elected to maintain the dog on cyclosporine (2.5 mg/kg [1.1 mg/lb, PO, q 12 h) indefinitely. Recurrent fistulas in the other 6 dogs were all located over an anal sac. Each of these dogs was again treated with cyclosporine for 3 to 8 weeks and then underwent bilateral anal sacculectomy and, if necessary, local excision of any remaining pinpoint fistulas. Surgical incisions were all closed primarily, and dogs were discharged from the hospital on the day of surgery. The surgical site did not require any specific aftercare, and none of the dogs developed complications. Histologic evaluation of the excised anal sac and duct tissue revealed mononuclear infiltration of the apocrine glad epithelium and, in 2 dogs, pigmentary incontinence of the anal sac epithelium. Four of the 6 dogs did not have any further evidence of recurrence between 11 and 13 months after surgery. One dog developed another fistula, which was excised. Histologic examination of the excised tissue revealed remnants of secretory anal sac epithelium. The other dog developed a fistula when it began an estrus cycle and was successfully treated with an additional course of cyclosporine and ovariohysterectomy. The dog did not have any further evidence of recurrence for 10 months after cyclosporine administration was again discontinued.
We did not detect any differences in regard to sex, age, previous treatment, status of anal sacs on digital palpation, or severity of disease between dogs that had a recurrence of fistulas and those that did not. However, mean duration of fistulas before inclusion in the study was significantly longer for dogs that developed recurrence (12.8 months; range, 1 to 36 months) than for dogs that did not (5.1 months; range 1 to 18 months).
Results of digital palpation – On initial digital palpation, the anus and distal portion of the rectum adjacent to the fistulas were firm, irregular, and thickened. Even after dogs were sedated, the swelling and thickening of the perianal tissues, along with the associated signs of pain and discomfort, often hindered thorough digital palpation. At the initial evaluation, the anal sacs were usually indiscernible.
After 16 weeks of cyclosporine treatment, most dogs did not require sedation to allow examination and palpation of the perianal region. The anal and rectal mucosa was smooth and soft. The anal sacs in most dogs were still indiscernible or could be palpated as firm, empty, pea-sized nodules. The anal sacs could be identified as intact, fluid-filled sacs bilaterally in only 4 dogs and unilaterally in only 1 dog. In 2 dogs in which the anal sacs were palpable bilaterally, 1 anal sac was full distended by inexpressible, possibly because of stricture of the anal sac duct associated with healing of the fistulas. Therefore, anal sacculectomy was recommended.
Blood cyclosporine concentration – Mean initial dosage of cyclosporine was 5.5 mg/kg, PO, every 12 hours (2.5 mg/lb; range, 4.8 to 6.8 mg/kg [2.2 to 3.1 mg/lb]). Mean ± SD trough blood concentration was 506 ± 307 ng/ml (range, 200 to 1,135 ng/ml). Mean adjusted dose was 5.9 mg/kg, PO, every 12 hours (2.7 mg/lb; range, 2.7 to 8.0 mg/kg [1.2 to 3.6 mg/lb]), and mean ± SD trough blood concentration was 488 ± 157 ng/ml (range, 200 to 870 ng/ml).
Adverse effects - The most frequently reported adverse effect associated with cyclosporine administration was increased shedding of hair. Shedding was most noticeable early in the course of treatment, and, by 16 weeks, the old coat had typically been replaced by a softer, glossier coat. One dog, a long-haired German Shepherd Dog, developed hypertrichosis during treatment. Owners of 2 dogs that had chronic generalized dermatitis and otitis externa reported a marked improvement in the skin condition during cyclosporine treatment. Vomiting was reported by owners of 5 of the 10 dogs receiving cyclosporine and cephalexin concurrently. Vomiting resolved when administration of cephalexin was discontinued.
Four dogs developed transient lameness during cyclosporine treatment. Two of these dogs has a history of panosteitis when they were younger; however, radiographs obtained when they were lame while receiving cyclosporine did not reveal any diagnostic lesions. The lameness resolved in 3 dogs while they were still receiving cyclosporine, but in 1 dog, the lameness resolved only after cessation of cyclosporine treatment. Opportunistic infection was not a problem in any of the dogs receiving cyclosporine except 1 dog that developed pyometra. Administration of cyclosporine was discontinued while the dog was treated for pyometra but was resumed thereafter. Serum creatinine concentration remained within reference limits in all dogs during cyclosporine treatment.
Discussion
A randomized controlled trial is the best method for assessing efficacy of a particular treatment.17 The first 4 weeks of this study were conducted as a randomized controlled trial and demonstrated that cyclosporine was effective in the treatment of perianal fistulas. The remainder of the study, although, in essence, a case series, provided valuable data on the longer term outcome of this treatment.
The 4-week duration for this trial was chosen on the basis of the expected progression of the disease in dogs and previously reported results of cyclosporine treatment in a series of dogs with perianal fistulas.16 Four weeks was thought to be sufficient time to determine whether cyclosporine would be effective without being so long as to permit the fistulas to become much worse because of a lack of beneficial treatment. In fact, substantial improvement was evident within 2 weeks in dogs that received cyclosporine, although full healing required a longer period. The initial improvement reported by a few owners of control-group dogs was equivocal and not sustained and may have been a result of cephalexin treatment or a placebo effect.
The 16-week duration of cyclosporine treatment was chosen on the basis of a previous observation that the likelihood of recurrence was greater in dogs treated for only 8 weeks.16 To minimize variability, duration of treatment was 16 weeks for all dogs in the study. In a clinical setting, it may be more appropriate to administer cyclosporine for as long as there is progressive improvement and for 4 additional weeks after all fistulas appear healed.
Cyclosporine is a potent immunosuppressive drug 16-20; however, there is tremendous Inter- and intrapatient variation in the rate of absorption and metabolism of cyclosporine in dogs and human beings. Thus, blood concentrations must be carefully monitored to ensure that they remain within the therapeutic range.21 A new microemulsion formulation of cyclosporine?, which has greater bioavailability and less variability in pharmacokinetic parameters, may enable more accurate targeting of blood concentrations.22,23 Various drugs interfere with the kinetics of cyclosporine, and cyclosporine can interfere with the metabolism of various other drugs.22,24 The target range for trough blood cyclosporine concentrations used in this study (400 to 600 ng/ml) was selected on the basis of the reported therapeutic range for cyclosporine when cyclosporine is used as a part of the immunosuppressive protocol for dogs with renal allografts.24 In some dogs, the cyclosporine dosage was increased, despite obvious clinical improvement at the lower dosage, to maintain trough blood concentrations within the target range. In our experience with more than 50 subsequent dogs with perianal fistulas, lower dosages of cyclosporine (1.75 to 3 mg/kg [0.8 to 1.4 mg/lb], PO, q 12 h), which resulted in blood concentrations between 100 and 300 ng/ml, were also effective. Thus, in a clinical setting, the decision to increase cyclosporine dosage need not be made solely on the basis of trough blood concentrations but also on the basis of response to treatment. Further study is needed to determine the optimal cyclosporine dosage.
In this study, blood cyclosporine concentrations were determined by use of a monoclonal radioimmunoassay6 that specifically measured only the parent cyclosporine compound. Other assays can cross-react with cyclosporine metabolites and may indicate higher concentrations. It is important that blood samples and not serum samples, be used to determine the cyclosporine concentrations.
Hirsutism and hypertrichosis are recognized adverse effects of cyclosporine in human beings 15-16, and hypertrichosis was reported for 1 dog in this study. Excessive shedding, reported for several other dogs in the study, was likely attributable to the same mechanism. Nephrotoxicosis and hepatotoxicosis have been reported to develop in some human beings receiving cyclosporine18 but were not evident in these dogs. Seizures have been associated with cyclosporine administration in cats,4 but such seizures have not reportedly developed in dogs and were not reported in this study. Development of pyometra in 1 dog in this study may or may not have been associated with cyclosporine treatment, but it underscores the importance of being vigilant for an concurrent infectious process prior to and during cyclosporine treatment. The cause of the lameness in 4 dogs was not determined and deserves further investigation.
Cephalexin was administered to all dogs in this study because of concerns that the immunosuppressive effects of cyclosporine would exacerbate any infection associated with the fistulas. However, results of this study and a previous report16 suggest that administration of cephalexin is not necessary to achieve an effective response with cyclosporine.
Even though 7 dogs in this study eventually required surgery, the surgical procedures were less extensive than would have been required if the dogs had not been treated with cyclosporine, and the risk of postoperative complications (eg. fecal incontinence, anal stricture) was, therefore, lower. In addition, postoperative management was less extensive. Thus, cyclosporine administration appeared to be beneficial even in those dogs in which the fistulas were not completely healed with cyclosporine treatment alone.
The cause of perianal fistulas remains unknown, but the response of dogs in this study to cyclosporine treatment and the response of dogs in a previous study14 to prednisone would support an immune-mediated role. In addition, recurrent fistulas in dogs in this study were all associated with anal sacs, suggesting that these structures also play a role.
Intuitively, it would seem that if perianal fistulas are immune-mediated, the immunosuppressive treatment will be needed continuously or intermittently throughout the dog's life. However, 11 dogs in this study that were treated with cyclosporine alone did not have any evidence of recurrence after cyclosporine administration was discontinued. Admittedly the follow-up period was only 12 to 19 months and long-term follow-up information is needed to determine whether intermittent treatment will be necessary. In this study, dogs that did not develop a recurrence had had fistulas for a shorter period prior to cyclosporine treatment than did dogs that did develop a recurrence, suggesting that cyclosporine treatment should be initiated as early as possible.
