SEARCH

SEARCH BY CITATION

Keywords:

  • colostomy;
  • complication;
  • radiotherapy;
  • rectovaginal fistula;
  • treatment

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

Analysis of the clinical course of patients with postirradiation rectovaginal fistula after fecal diversion. The studied group included 17 women with postirradiation rectovaginal fistula who underwent fecal diversion as a sole mode of treatment, between January 1987 and December 2002, in our department. All patients were subjected to radiotherapy due to cancer of the uterine cervix, administered 5–107 months before the fistula appearance (mean, 22.9 months). In 3 of 17 patients (18%), spontaneous closure of fistula was observed after 5, 6, and 9 months, respectively, from fecal diversion. Closure was confirmed by endoscopy. Length of follow-up after fecal diversion ranged from 0.5 to 122 months. The actuarial probability of spontaneous closure of postradiotherapy rectovaginal fistula was 0.24 at 9 months of follow-up and then remained stable thereafter. In conclusion, colostomy alone gives hardly a chance for closure of the postradiotherapy rectovaginal fistula. Additional surgical measures are necessary.

Rectovaginal fistula in patients irradiated for cancer of the uterine cervix is a rare complication, especially when modern radiotherapy techniques are employed(1–4). Despite the low incidence, it causes significant emotional, social, and sexual problems for the woman, as well as a substantial distress for the treating oncologist. The most frequent patient’s complaints include vaginal passage of flatus and leakage of stool from the rectum to vagina. The repair is very difficult because of damage of tissues caused by radiotherapy and, in case of successful exclusive irradiation, remission of tumor. In consequence, spontaneous healing of extensive defects of normal tissues initially infiltrated by the tumor is rare. Attempts of surgical fistula repair have been undertaken with various results. The methods of repair include colon resection, “pull through” procedures, and interposition of well-vascularized flaps(5–11). However, in many patients, permanent diverting colostomy is the only possible treatment(12,13). Moreover, a temporary colostomy is often suggested before an attempt of fistula repair in order to decrease the sepsis and edema of distal bowel(6,10).

Despite extensive search of the current literature and Internet medical databases, we found little information concerning a probability of spontaneous closure of rectovaginal fistula after the fecal diversion in patients who had undergone radiotherapy for cervical cancer(12). That prompted us to analyze our material and assess the probability of such an event.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

Between January 1987 and December 2002, 21 women with iatrogenic, postirradiation rectovaginal fistulas underwent fecal diversion as a sole mode of treatment at the Department of Surgical Oncology, Medical University of Lodz, Poland. All these patients were subjected to radiotherapy due to cancer of the uterine cervix. Retrospective review of the patients’ files identified 17 eligible women whose clinical data were complete. These patients’ data are the data for this study. Four patients were excluded because of incomplete records.

Patients’ characteristics

Seven patients underwent surgery as a primary treatment of cervical cancer: six underwent total hysterectomy and one underwent hysterectomy. In these women, radiotherapy was given as adjuvant postoperative treatment (7/17, 41%). Ten patients underwent radical radiotherapy as a primary treatment (10/17, 59%). No patient received chemotherapy, and no patient suffered from concomitant inflammatory bowel disease.

Each patient underwent combined modality irradiation including external beam radiotherapy followed by intracavitary brachytherapy. External beam irradiation was delivered with the use of Co-60 unit. Median dose was 40 Gy (range, 38.25–50.00 Gy; mean, 40.9 Gy); 20 fractions were used. Uterovaginal and vaginal brachytherapy were performed in eight and nine patients, respectively (uterovaginal treatment was done in eight of ten cases of radical irradiation). In three cases, high–dose rate brachytherapy was used (3 × 6 Gy). In the remaining 14 cases, radium-226, low–dose rate brachytherapy was employed up to the median of 30.6 mRa (range, 24–32 mRa; mean, 27.7 mRa). The dose applied to uterus ranged from 20.8 to 24 mRa (median, 24 mRa; mean, 23.7 mRa) (Table 1).

Table 1.  Characteristics of 17 patients with radiation-induced rectovaginal fistulas
NumberDate of cancer diagnosisAge of patient at diagnosis (years)Surgery typeRadiotherapyTime from the end of radiotherapy to fistula formation (months)
Teleradiotherapy (Co-60 [Gy])Brachytherapy
Vagina HDR (Gy)Vagina Rad (mRa)Uterus Rad (mRa)
  1. HDR, high-dose rate; Rad, radium.

1198243Not performed4003021107
2198469Panhysterectomy40032025
3198439Not performed400322436
4198558Not performed40031.220.842.5
5198662Not performed400322413
6198641Not performed400322417
7198759Panhysterectomy4003009
8198758Panhysterectomy40024026
9198954Not performed400322419
10199067Panhysterectomy38.25028.805
11199058Not performed40032249
12199244Not performed38.25032287
13199444Panhysterectomy38.2503009
14199741Not performed403 × 60019
15199843Hysterectomy443 × 60010
16199850Panhysterectomy503 × 60012
17200043Not performed4600012

In each case, diagnosis of rectovaginal fistula was made on the basis of clinical signs and symptoms. In each case, the diagnosis was confirmed by colposcopy and rectoscopy. In all patients, radiation-induced fistulas were spontaneous (ie, they were not caused by surgical procedures). Median time between the end of radiotherapy and the diagnosis of rectovaginal fistula was 15 months (range, 5–107 months; mean, 22.9 months). In all patients, recurrent disease at the time of fistula diagnosis was excluded through clinical and radiologic evaluation. In all patients, loop colostomy was made. Median age of patients at the time of fecal diversion was 52 years (range, 42–72 years; mean, 53.6 year). Spontaneous closure of the rectovaginal fistula was confirmed by rectoscopy and colposcopy (no routine abdominopelvic x-ray examination was scheduled).

Statistical analysis

To evaluate the results of fecal diversion in patients with postradiotherapy rectovaginal fistula, we calculated the absolute rate of spontaneous closure of fistula after surgery. We also calculated the probability of spontaneous closure of fistula according to the Kaplan–Meier method for estimation of survival. At the first step, we draw a curve of the probability of living with patent fistula. Spontaneous closure of fistula was considered for completed data. Probability was calculated from time of surgery to fistula’s closure or to the last visit in case of the disease-free patients or to the other events (diagnosis of relapse, an additional fistula or the last follow-up visit without progression of the disease). As the final step, probability of spontaneous closure of fistula was calculated ([probability of spontaneous closure] = 1 − [probability of living with patent fistula]).

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

In 3 of 17 patients (18%), spontaneous closure of fistula was observed after 5, 6, and 9 months, respectively, from fecal diversion. Length of follow-up after the fecal diversion ranged from 0.5 to 122 months. Each of these three women had received radical radiotherapy before rectovaginal fistula developed (in subgroup of patients who underwent hysterectomy/total hysterectomy and adjuvant radiotherapy before the development of rectovaginal fistula, no spontaneous closure of fistula was observed).

In the remaining 14 patients, spontaneous closure of rectovaginal fistula did not occur (14/17; 82%) during the follow-up period. In four of them, relapse of the cancer was diagnosed (local recurrence in two patients; local recurrence and distant metastases in one patient; and distant metastases in one patient). Six patients from this group who developed additional fistula were as follows: between vagina and urinary bladder (four patients) and between vagina and small intestine (two patients). In the remaining four patients, a closure of fistula was not observed during the follow-up period.

The actuarial probability of spontaneous closure of postradiotherapy rectovaginal fistula was 0.24 at the 9 month of follow-up and then remained stable thereafter (Fig. 1).

image

Figure 1. The actuarial probability of living with patent fistula (in 17 patients with postirradiation rectovaginal fistula after the fecal diversion).

Download figure to PowerPoint

Three women in whom spontaneous healing of rectovaginal fistula occurred were subjected to further follow-up. In each case, confirmation of the closure of fistula (abdominopelvic x-ray with contrast medium) and the restoring of physiologic passage of stools were planned after 4–6 months, after the diagnosis of spontaneous closure of rectovaginal fistula. However, during follow-up period after the closure of fistula, one patient developed distant metastases, and the other one developed an additional fistula between the urinary bladder and the vagina. The third of these three patients refused surgical reversal of colostomy. She explained that she felt comfortable and safe with colostomy. Therefore, no attempt was made to close the colostomy after the fistula closure.

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References

Spontaneous closure of fistula was observed in 18% (3/17) of our patients. An absolute rate does not take into account factors such as the length of patient follow-up. In patients whose data are censored for various reasons, spontaneous closure of rectovaginal fistula might have occurred or will occur in the future. Therefore, to assess the effectiveness of fecal diversion, we also used the Kaplan–Meier method in order to calculate the probability of spontaneous closure of the fistula. The actuarial probability of spontaneous closure of postradiotherapy rectovaginal fistula was 0.24 at 9 months after the diversion and then remained at the same level. However, the difference between the absolute rate (18%) and the actuarial probability (0.24) was small. Importantly, in 7 of 17 patients (41%), subsequent fistulas occurred (in one patient after the spontaneous closure of the rectovaginal fistula, whereas in other six patients—two fistulas were present). This underlines the progressive evolution of the severe late radiation injury(14,15). A large study of radiotherapy complications after postoperative radiotherapy for endometrial cancer showed the reversible character of moderate late effects, while severe injury hardly regress spontaneously and tend to worsen with follow-up duration(14). Small sample and incomplete data concerning the sizes of each fistula precluded the analysis of relationship between the probability of spontaneous closure of the fistula and its size. However, in all three cases in which spontaneous healing took place, fistulas were small. The diameter of fistula in each case was smaller than 8 mm. The closure of fistula in all three cases was not confirmed with the use of x-ray. This procedure was planned before an attempt of colostomy closure. Unfortunately, in two patients, progression of cervical cancer occurred and one patient refused further treatment. The number of rectovaginal fistulas treated surgically in our department seems to be high. All our patients were referred from the Regional Center of Oncology in Lodz (central Poland), where they underwent radiotherapy. The Regional Center of Oncology in Lodz is a large oncology hospital covering an area of a population of approximately 3,000,000 people. The incidence and the mortality of cervical cancer in Poland is still very high (13/100,000 and 7/100,000, respectively). High percentage of patients presented with advanced disease for which curative radiotherapy or surgery followed by adjuvant irradiation are proposed. Indeed, mortality due to cervical cancer in Poland was the highest among 17 European countries taking part in the Eurocare-3 study(16). Therefore, in the period of 15 years, such a relatively high number of patients were referred to our department, as the reference surgery center for the surgical treatment of the radiation-induced rectovaginal fistulas. We did not analyze the incidence of fistulas in the patients treated with radiotherapy for gynecological cancer; however, the rates of severe postradiotherapy bowel complications of up to 8% were reported by other authors(14,17–19). Very little information is available in the modern medical literature concerning the chance of spontaneous cure of postradiotherapy rectovaginal fistula. Segreti et al.(12) described a group of 47 patients with rectovaginal fistula in whom loop colostomy was used for fecal diversion. Only 83% of cases underwent prior pelvic irradiation. In four patients, spontaneous closure of rectovaginal fistula occurred, and consequently, loop colostomy reversal procedures were undertaken. Finally, only one closure was successful; however, this patient had not been irradiated. Her fistula was secondary to surgery for colon cancer.

Although there is general agreement that end colostomy is superior to loop colostomy as a method of fecal diversion, loop colostomy is commonly employed(12,20–22). As no attempt of colostomy reversal was made in our patients with spontaneously closed fistula, the advantage of simple extraperitoneal loop colostomy closure was not realized. Therefore, when choosing the method of fecal diversion in future, we should keep this fact in our minds.

Our three patients, in whom spontaneous closure of rectovaginal fistula occurred, described that fact as a benefit. Despite the inconvenience of having the colostomy, the women were satisfied with the decrease of unpleasant mucinous discharge from vagina after the closure of the fistula. Such observation was previously reported by others(9). Importantly, in our series spontaneous fistula closure occurred exclusively in patients who underwent exclusive radiotherapy (not proceeded by pelvic surgery), and no closure was observed in patients treated with surgery followed by radiotherapy. One can hypothesize that intact pelvic anatomy helps the tissue repair after irradiation, even though higher doses are delivered in case of curative radiotherapy. Moreover, different etiology of the fistula in the operated patients and in the patients treated with radiotherapy with present tumor may be suspected. In case of exclusive radiotherapy, the fistula may occur due to complete regression of the tumor infiltrating the surrounding organs. The mechanisms of tissue defect repair probably differ between these two patient populations. Already some authors reported higher risk of radiation-induced complications in patients who underwent preirradiation pelvic surgery(23–25). Based on our observations, we can propose the hypothesis that such combination (surgery followed by radiotherapy) makes also the healing of these complications less probable. Obviously, such hypothesis must be verified in a larger patient series.

Attempts of surgical fistula repair have been undertaken by other authors with much better results than presented in this study. Park’s coloanal sleeve anastomosis was successfully used by Nowacki(6). Author reported good functional results in 18 of 23 operated women. White et al.(7) and Aartsen and Sindram(9) repaired radiation-induced rectovaginal fistulas with the interposition of the bulbocavernosus muscle (Martius procedure). The success rates in their material were 11 of 12 and 13 of 14, respectively.

In conclusion, the goal of colostomy is to provide complete diversion of the stools and to prevent the patient from the passage of stool through vagina. Colostomy alone gives hardly a chance of closing the rectovaginal fistula. Additional surgical measures are necessary.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. References
  • 1
    Ogino I, Kitamura T, Okamoto N et al. Late rectal complication following high dose rate intracavitary brachytherapy in cancer of the cervix. Int J Radiat Oncol Biol Phys 1995;31:72534.
  • 2
    Kutzner J, Knappstein T, Hager S, Koch H. Results of radiation therapy in cervix carcinoma with reference to side effects. Strahlenther Onkol 1886;162:54954.
  • 3
    Alert J, Jimenez J, Beldarrain L, Montalvo J, Roca C. Complications from irradiation of carcinoma of the uterine cervix. Acta Radiol Oncol 1980;19:135.
  • 4
    Szawlowski A. Popromienna przetoka pochwowo-odbytnicza (fistula rectovaginalis postradiologica): problem kliniczny. Nowotwory 1984;34:21321.
  • 5
    Lucarotti ME, Mountford RA, Bartolo DC. Surgical management of intestinal radiation injury. Dis Colon Rectum 1991;34:8659.
  • 6
    Nowacki MP. Ten years of experience with Park’s coloanal sleeve anastomosis for the treatment of post-irradiation rectovaginal fistula. Eur J Surg Oncol 1991;17:5636.
  • 7
    White AJ, Buchsbaum HJ, Blythe JG, Lifshitz S. Use of bulbocavernosus muscle (Martius procedure) for repair of radiation-induced rectovaginal fistulas. Obstet Gynecol 1982;60:1148.
  • 8
    Nowacki MP, Szawlowski AW, Borkowski A. Park’s coloanal sleeve anastomosis for treatment of postirradiation rectovaginal fistula. Dis Colon Rectum 1986;29:81720.
  • 9
    Aartsen EJ, Sindram IS. Repair of the radiation induced rectovaginal fistulas without or with interposition of the bulbocavernosus muscle (Martius procedure). Eur J Surg Oncol 1988;14:1717.
  • 10
    Borkowski A, Nowacki M. Simultaneous repair of post-irradiation vesicovaginal and rectovaginal fistulas. J Urol 1982;128:9268.
  • 11
    Parks AG, Allen CLO, Frank JD, McPartlin JF. A method of treating post-irradiation rectovaginal fistulas. Br J Surg 1978;65:41721.
  • 12
    Segreti EM, Levenback C, Morris M, Lucas KR, Gershenson DM, Burke TW. A comparison of end and loop colostomy for fecal diversion in gynecologic patients with colonic fistulas. Gynecol Oncol 1996;60:4953.
  • 13
    Kimose HH, Fischer L, Spjeldnaes N, Wara P. Late radiation injury of the colon and rectum—surgical management and outcome. Dis Colon Rectum 1989;32:6849.
  • 14
    Jereczek-Fossa B, Jassem J, Nowak R, Badzio A. Late complications after postoperative radiotherapy in endometrial cancer: analysis of 317 consecutive cases with application of linear-quadratic model. Int J Radiat Oncol Biol Phys 1998;41:32938.
  • 15
    Pedersen D, Bentzen SM, Overgaard J. Reporting radiotherapeutic complications in patients with uterine cervical cancer. The importance of latency and classification system. Radiother Oncol 1993;28:13441.
  • 16
    Coleman MP, Gatta G, Verdecchia A et al.; EUROCARE Working Group. EUROCARE-3 summary: cancer survival in Europe at the end of the 20th century. Ann Oncol 2003;14(Suppl. 5):v12849.
  • 17
    Potish RA, Dusenbery KE. Enteric morbidity of postoperative pelvic external beam and brachytherapy for uterine cancer. Int J Radiat Oncol Biol Phys 1990;18:100510.
  • 18
    Randall ME, Wilder J, Greven K, Raben M. Role of intracavitary cuff boost after adjuvant external irradiation in early endometrial carcinoma. Int J Radiat Oncol Biol Phys 1990;19:4954.
  • 19
    Stryker JA, Podczaski E, Kaminski P, Velkley DE. Adjuvant external beam therapy for pathologic stage I and occult stage II endometrial carcinoma. Cancer 1991;67:28729.
  • 20
    Fontes B, Fontes W, Utiyama EM, Birolini D. The efficacy of loop colostomy for complete fecal diversion. Dis Colon Rectum 1988;31:298302.
  • 21
    Palmer JA, Bush RS. Radiation injuries to the bowel associated with the treatment of carcinoma of the cervix. Surgery 1976;80:45864.
  • 22
    Hallfeldt K, Schmidbauer S, Trupka A. Laparoscopic loop colostomy and advanced ovarian cancer, rectal cancer and rectovaginal fistulas. Gynecol Oncol 2000;76:3802.
  • 23
    Greven KM, Corn BW. Endometrial cancer. Curr Probl Cancer 1997;21:65127.
  • 24
    Hanks GE, Herring DF, Kramer S. Patterns of care outcome studies. Results of the national practice in cancer of the cervix. Cancer 1983;51:95967.
  • 25
    Lanciano RM, Martz K, Montana GS, Hanks GE. Influence of age, prior abdominal surgery, fraction size, and dose on complications after radiation therapy for squamous cell cancer of the uterine cervix: a pattern of care study. Cancer 1992;69:212430.