- Email: [email protected]
Chylous ascites after whole-abdomen irradiation for gynecologic malignancy
In!. J Radiamn Oncolo~1. Bid Php.. Vol. Printed in the U.S.A. All n&s reserved.
19. pp. 435-438 Copyright
0360.3016/90 $3.00 t .IKI 0 1990 Pergamon Press plc
??Brief Communication
CHYLOUS ASCITES AFTER WHOLE-ABDOMEN IRRADIATION FOR GYNECOLOGIC MALIGNANCY SAMUEL S. LENTZ, M.D.,*
MARK F. SCHRAY, M.D.+ AND TIMOTHY
0. WILSON, M.D.*
Mayo Clinic and Mayo Foundation, Rochester, MN 55905 From November 1!)81 through December 1987, 207 patients received whole-abdomen irradiation (WAI) for gynecologic malignancies at the Mayo Clinic. In seven (3%) of these patients, chylous ascites subsequently developed; one additional patient with chylous ascites after WA1 for a gynecologic malignancy was referred to us from another institution. In these eight patients, irradiation was done either adjuvantly (five patients) or as salvage therapy after chemotherapy failure (three patients). Chylous ascites was confirmed by laboratory analysis in six cases and was presumed based on the clinical course in two cases. Mean cumulative radiation doses were 2,925 and 5,122 cGy to the abdomen anld pelvis, respectively, with para-aortic boosts administered in six cases to a mean cumulative dose of approximat’ely 4,200 cGy. The mean time from completion of WA1 to development of ascites was 12 months (range, 6 to 18 months). In six patients, therapy was conservative-observation and diuretics. Two other patients required multiple paracenteses for relief of abdominal distention. Parenteral nutrition was given to two patients who had associated1 radiation enteritis. The ascites resolved in all eight cases at a mean of 18 months (range, 8 to 30 months) after development. At a mean follow-up of 57 months after initial diagnosis and 16 months after resolution of the as’cites, seven patients are without evidence of disease and one patient died of recurrent carcinoma. Distinguishing this clinical entity from recurrent carcinoma is important because of its benign course and its resolution with conservative management. Ovarian cancer, Chylous ascites, Whole-abdomen
radiation therapy.
INTRODUCTION
veloping after whole-abdomen irradiation (WAI) for primary gynecologic malignancies. This report analyzes the time course, natural history, and management of this entity in our experience.
Chylous ascites is a peritoneal effusion, high in fat and protein contents, that results from lymphatic obstruction. The incidence of chylous ascites is difficult to ascertain but recently has been reported to be approximately 1 case per 12,000 admissions at the Massachusetts General Hospital from 1977 to 1982 ( 15). A study by Rovelstad et al. (16) at the Mayo Clinic noted that, in 116 consecutive cases of ascites requiring paracentesis, chylous ascites was found in 10. The causes of chylous ascites have been classified (8) as spontaneous or traumatic. In children most cases are related to congenital lymphatic abnormalities whereas in adults the cause is neoplasm in 80% of cases. Retroperitoneal, mesenteric, or small bowel fibrosis secondary to irradiation or some other influence is an established cause of chylous ascites. Sipes et al. (17) reported two cases of chylous ascites as a sequel of pelvic radiation therapy, and Murray et al. ( 13) described three cases after whole-abdomen radiotherapy for ovarian cancer. We have observed ejght cases of chylous ascites de-
METHODS
Patients and radiation treatments Of the 207 patients who received WA1 for gynecologic malignancy at the Mayo Clinic from 198 1 through 1987, post-irradiation ascites had developed in 7 (3.4%). One additional patient with chylous ascites after adjuvant WA1 for a gynecologic malignancy was referred to us from another institution. The characteristics of these eight patients are presented in Table 1. The mean age was 66 years (range, 47 to 7 1 years); the mean body weight at the initiation of radiation therapy was 68 kg (range, 46 to 88 kg). The mean number of pretreatment operations was 1.5 per patient: two patients had two abdominal explorations, and one patient had
Presented at the Seventeenth Annual Meeting of the Western Association of Gynecologic Oncologists, San Francisco, CA, 1720 May 1989. * Department
of Obstetrizs
AND MATERIALS
+Division of Radiation Oncology. Reprint requests to: Timothy 0. Wilson, M.D., Mayo Clinic, 200 First Street SW, Rochester, MN 55905. Accepted for publication 22 February 1990.
and Gynecology.
435
1. J. Radiation
436
Oncology
0
Biology0 Physics
August
1990. Volume
Patient
Table 1. Characteristics of the eight cases
Radiation
19. Number
Davebpment
Resolution
Length of follow-up
of ascites
of ascites
after rewlution
therapy
Characteristic
Patients, no.
Age, yr 160 >60 Weight, kg 550 >50 Pre-irradiation laparotomy, no. 1 2 3 Para-aortic node dissection Pelvic node dissection
2 6 3 5
Site of
Ovary Ovary
Endometrium
Carcinosarcoma
* After chemotherapy failure. + Received WA1 elsewhere.
z
\NEc!4rnO ANED16nw
.
/+NED~~JM
6-
.
j,+
NED 21 mo
7-
.
//+
NED 24 ma
Z /&ANED29mo I
I
I
18
12
6
0
I
I
I
I
I
I
6
12
la
24
30
36
Fig. 1. Chronologic sequence of events. 0, irradiation; 0, resolution of ascites; A, last follow-up; DOD, dead of disease; NED, no evidence of disease.
Table 2. Characteristics of neoplasms
Serous, grade 3 Serous, grade 4 Serous, grade 2 Mutinous, grade 2 Mutinous, grade 3 Endometrioid, grade 3 Adenosquamous, grade 3
. .
Months
The mean time from completion of WA1 to development of ascites was 12 months (range, 6 to 18 months)
Ovary Ovary Ovary Ovary+ Ovary
3d-
0-a
RESULTS
Histology
-w-‘zmo
5-
three. Pretreatment para-aortic node dissection, performed in three patients, revealed a grossly positive 7-cm mass in one patient; it was completely excised. A pelvic lymphadenectomy was performed in the one patient with endometrial cancer. In the seven cases treated at the Mayo Clinic, WA1 was administered by an open-field technique as described (11). Ovarian malignancies were present in 7 of the 8 cases (Table 2). WA1 was used as salvage treatment after chemotherapy failure in three patients and as primary adjuvant therapy in four cases. The eighth patient received primary adjuvant WA1 for endometrial cancer. Mean cumulative radiation doses, given in 150-cGy daily fractions, were 2,925 and 5,122 cGy to the abdomen and pelvis, respectively, with para-aortic boosts administered in six cases to a mean cumulative dose of 4,200 cGy. There was no therapy in the interval between administration of the radiation and development of chylous ascites.
primary
l-1
2
Stage
Reason for WA1
IV III III I IC
Salvage* Salvage* Salvage* Adjuvant Adjuvant
III
Adjuvant
IIC
Adjuvant
IC
Adjuvant
(Fig. 1). Although some patients were asymptomatic at presentation, most had symptoms including abdominal bloating, dyspepsia, nausea, and vomiting. Examination generally revealed increased abdominal girth, distention, and a fluid wave. CT scans, obtained in five cases, all revealed ascites. Paracentesis was the primary diagnostic modality in seven cases. Peritoneoscopy also was performed in three of these cases. Laparotomy was diagnostic in one patient who had a concurrent small bowel obstruction. The ascitic fluid had a milky appearance in six cases and was serous appearing in two cases. Lipid analysis was performed in six cases (five chylous appearing and one serous) and the results are shown in Table 3. Chylomicrons were the major structural component of the ascitic lipids. In one case of chylous-appearing ascites and one case of serous-appearing ascites, lipid analysis was not performed and the diagnosis of chylous ascites was based on clinical features. Cytologic examination was negative for malignancy in all eight cases. Conservative therapy, consisting of observation and potassium-sparing diuretics, was used as the sole management in six patients. Two other patients required multiple paracenteses for symptomatic relief of abdominal distention. In addition, parenteral nutrition was used for a brief period in two patients as part of the management of associated radiation enteritis. There was no temporal relationship between parenteral nutrition and resolution of the ascites in these two patients. The ascites resolved clinically in all eight cases (CT scan confirmation in five) at a mean of 18 months (range, 8 to 30 months) after its development. At a mean follow-up of 57 months after initial diagnosis and a mean interval of 17 months (range, 3 to 29 months) after resolution of ascites, seven patients Table 3. Lipid analysis of ascitic fluid Test
Mean
Range
Specific gravity Total protein, g/dL Triglyceride, mg/dL Cholesterol, mg/dL
1.022 4.99 1,069 121
1.018-1.025 4.01-6.64 229-2,568 70-203
Chylous ascites after irradiation 0 S. S. LENTZ et al.
are without evidence of disease and one patient has died of recurrent carcinoma. DISCUSSION The diagnosis of ascites generally can be made clinically in conjunction with radiographic techniques if necessary. Paracentesis is confirmatory, allowing diagnostic analysis of the fluid. The fact that the fluid was not chylous in appearance in all of the present cases is not contradictory. Rovelstad et al.(16) found that, among 20 patients with chylous ascites secondary to cancer who had total serum lipids > 350 mg/dL, only 8 had grossly chylous fluid. Thus, exudation of lymph with a relatively high fat content into the peritoneal cavity may occur without necessarily producing fluid with a chylous appearance. The results of the laboratory analyses in .the present series are consistent with those defining chylous ascites in the literature (10). Lymphangiography is of little diagnostic benefit and, in fact, has been associated with a transient exacerbation of the ascites, presumably secondary to further obstruction of compromised lymphatics by the contrast agent-induced inflammatory reaction ( 18). The mean time of 12 months from completion of irradiation to the diagnosis of ascites supports the presumption of radiation injury as the etiologic factor in the development of the chylous ascites. It is postulated, although not specifically documented in the present cases, that radiation-induced fibrosis of the lymphatic vessels within the small bowel and its associated mesentery resulted in occlusion and subsequent extravasation of the chylous fluid. Although a retroperitoneal fibrotic process could be responsible, case reports in the literature support our supposition. Hurst and Edwards (7) described a case of chylous ascites developing after abdominal radiotherapy for recurrent colon carcinoma. At exploration, radiation damage to the small bowel was noted and chyle was leaking from the most severely affected portion. Intraoperative mesenteric lymphangiography revealed lymphatic obstruction in the mesentery. The ch.ylous ascites resolved after resection of the worst-affected segment of small bowel. Similar findings were noted in the three cases reported by Murray and Massey (13) Dilated lymphatics were seen in the root of the small bowel mesentery at laparotomy, suggesting this as the point of obstruction. The possible contribution of a para-aortic boost to the development of chylous ascites could be related to its effect on the base of the small bowel mesentery which is relatively immobile. In addilion, the report by Sipes et al. (17) of chylous ascites after pelvic irradiation also suggests an obstructive lymphatic process within the small bowel and its mesentery. Of interest, in investigating the lymphatic circulation, Blalock et al. (2) noted a lchylous effusion in only 3 of 74 animals undergoing over 250 operative procedures destructive to major lymphatic vessels. At autopsy, numerous lymphaticovenous anastomoses were noted. This
437
suggests that a retroperitoneal lymphatic injury, whether induced surgically or by radiation, rarely leads to chylous ascites. Rather, it is the effect on the intestinal lymphatics and associated lymphaticovenous collaterals that is responsible for the development of chylous ascites. It has been suggested that any process that leads to portal hypertension may predispose to the development of chylous ascites, including radiation-induced hepatitis (6). Even though mild abnormalities in liver function tests were noted in the present cases after irradiation, clinically significant liver dysfunction was not seen. No other consistent predisposing factors were identified in this group of eight cases in comparison with the 207 WA1 cases including a number of pretreatment abdominal operations and para-aortic lymph node dissections. Regarding the present study, the incidence of approximately 3% seen here may be higher than that suggested by case reports and may reflect the relatively large number of patients treated at the Mayo Clinic. In addition, when compared with the large experience at the Princess Margaret Hospital, we used a higher daily fraction size and mean cumulative dose to the entire abdomen (4). In addition, our frequent use of para-aortic boosts may be contributory (6 of the 8 present cases and nearly half of the total patient population receiving WAI). Various therapeutic modalities for chylous ascites have been described based on the above pathophysiology of the disease process (Table 4). Obviously, successful treatment of a neoplastic process that is obstructing efferent lymphatics would be important. Surgical intervention with ligation of traumatized lymphatic channels-for example, as seen after abdominal aortic aneurysm repair and truncal vagotomy-has been described (3, 14). In this situation, instillation of a lipophilic dye into the duodenum immediately before operation has been suggested to help locate the leak (9). Resection of the isolated segment of small intestine responsible for the chylous leakage had been advocated, as described above (7). The presumed diffuse intra-abdominal irradiation effect precluded surgical intervention as a therapeutic option in the cases presented here. Regarding symptomatic management, the use of peritoneovenous shunts, such as the LeVeen shunt, has been described as Table 4. Therapeutic modalities for chylous ascites Primary therapy Treatment for associated malignancy Surgical intervention Repair of leaking lymphatic channel Segmental small bowel resection Symptomatic therapy Peritoneovenous shunt Dietary alterations Medium-chain triglycerides (Tolerex) Parenteral nutrition Multiple paracenteses Diuretics Observation
438
1. J. Radiation Oncology 0 Biology 0 Physics
having generally poor results ( 15). Others have advocated exclusion of long-chain triglycerides from the diet with replacement by dietary supplements (i.e., Tolerex) consisting solely of short- and medium-chain triglycerides (5, 18). This is theoretically sound based on the known physiology of lipid metabolism; however, reported cases have not shown uniform success and the unpalatable nature of this diet is not acceptable to many patients. Total parenteral nutrition (TPN) has been used, thus eliminating enteral long-chain triglycerides ( 1, 12). The benefit from TPN probably is derived from its caloric and protein replacement rather than from the specific exclusion of LCT. The reported variable, and at times transient, response of chylous ascites to TPN generally is observed shortly after its institution. The brief use of parenteral nutrition in the two cases of radiation enteritis in this series was not related temporally to the resolution of the ascites and thus TPN was not considered to be a major therapeutic factor. Diagnostically, paracentesis allows cytologic examination to exclude recurrent neoplasms as well as lipid analysis to confirm the chylous nature of the ascites. In addition, paracentesis provides temporary relief of abdominal distention although it is associated with a small risk of infection and bowel injury. Cytologic examination to exclude recurrent neoplasm should be performed be-
August 1990. Volume 19, Number 2
cause the development of ascites 12 months after therapy suggests recurrence. Finally, medical management using diuretics such as spironolactone has frequently given variable subjective responses; however, it is difficult to evaluate its benefit over observation alone. In the present series, conservative therapy-consisting of observation, multiple paracenteses for symptomatic relief of abdominal distention, and diuretics-was used in all cases. Resolution was achieved in all cases without any obvious associated morbidity. Additionally, the development of chylous ascites was not a harbinger of recurrent disease in any case (based not only on negative cytologic examinations but also on the subsequent clinical course). Although one patient has died of recurrent carcinoma, the ascites had resolved without evidence of disease, as determined by exploratory laparotomy performed for bowel obstruction approximately 1 year prior to the recurrence. Resolution at a mean follow-up time of 18 months also is consistent with the establishment of collateral lymphatic drainage as well as lymphaticovenous anastomoses. The development of ascites is generally considered to be a manifestation of recurrent disease. With the increased use of WA1 as a treatment modality, chylous ascites as a self-limited treatment sequela should be considered in the differential diagnosis.
REFERENCES 1. Asch, M. J.; Sherman, N. J. Management of refractory chylous ascites by total parenteral nutrition. J. Pediatr. 95:260-
262; 1979. 2. Blalock, A.; Robinson, C. S.; Cunningham, R. S.; Gray, M. E. Experimental studies on lymphatic blockage. Arch. Surg. 34:1049-1071; 1937. 3. Bradham, R. R.; Gregorie, H. B.; Wilson, R. Chylous ascites
4. 5.
6.
7.
8. 9.
10.
following resection of an abdominal aortic aneurysm. Ann. Surg. 36:238-240; 1970. Dembo, A. J. Abdominopelvic radiotherapy in ovarian cancer: a IO-year experience. Cancer 55:2285-2290; 1985. Dharman, K.; Temes, S. P.; Wetherell, F. E.; Kendrick, M. J. Chyloperitoneum and chylothorax: a combined rare occurrence after retroperitoneal lymphadenectomy and radiotherapy for testis tumor. J. Urol. 13 1:346-347; 1984. Herz, J.; Shapiro, S. R.; Konrad, P.; Palmer, J. Chylous ascites following retroperitoneal lymphadenectomy: report of 2 cases with guidelines for diagnosis and treatment. Cancer 42:349-352; 1978. Hurst, P. A.; Edwards, J. M. Chylous ascites and obstructive lymphedema of the small bowel following abdominal radiotherapy. Br. J. Surg. 66:780-781; 1979. Kelley, M. L., Jr.; Butt, H. R. Chylous ascites: an analysis of its etiology. Gastroenterology 39:161-170; 1960. Lewis, J. W., Jr.; Storer, E. H. The management ofiatrogenic chylous ascites. Henry Ford Hosp. Med. J. 27: 140-144; 1979. Manten, H.; Barkin, J. S.; Rogers, A. I. Chylous ascites. Postgrad. Med. 7 1:79-84; 1982.
11. Martinez, A.; Schray, M. F.; Howes, A. E.; Bagshaw, M. A. Postoperative radiation therapy for epithelial ovarian cancer: the curative role based on a 24-year experience. J. Clin. Oncol. 3:901-911; 1985. 12. Meinke, A. H., III; Estes, N. C.; Ernst, C. B. Chylous ascites following abdominal aortic aneurysmectomy: management with total parenteral hyperalimentation. Ann. Surg. 190: 63 l-633; 1979. 13. Murray, J. M.; Massey, F. M. Chylous ascites after radiation therapy for ovarian cancer. Obstet. Gynecol. 44:749-751; 1974. 14. Musgrove, J. E. Post-vagotomy Ann. Surg. 175:67-69; 1972.
abdominal
chylous fistula.
15. Press, 0. W.; Press, N. 0.; Kaufman, S. D. Evaluation and management of chylous ascites. Ann. Intern. Med. 96:358364; 1982. 16. Rovelstad, R. A.; Bartholomew, L. G.; Cain, J. C.; McKenzie, B. F.; Soule, E. H. Ascites I. The value of examination of ascitic fluid and blood for lipids and for proteins by electrophoresis. Gastroenterology 34:436-450; 1958. 17. Sipes, S. L.; Newton, M.; Lurain, J. R. Chylous ascites: a sequel of pelvic radiation therapy. Obstet. Gynecol. 66:832835; 1985. 18. Weinstein, L. D.; Scanlon, G. T.; Hersh, T. Chylous ascites management with medium-chain triglycerides and exacerbation by lymphangiography. Am. J. Digest. Dis. 14:500509; 1969.
19. pp. 435-438 Copyright
0360.3016/90 $3.00 t .IKI 0 1990 Pergamon Press plc
??Brief Communication
CHYLOUS ASCITES AFTER WHOLE-ABDOMEN IRRADIATION FOR GYNECOLOGIC MALIGNANCY SAMUEL S. LENTZ, M.D.,*
MARK F. SCHRAY, M.D.+ AND TIMOTHY
0. WILSON, M.D.*
Mayo Clinic and Mayo Foundation, Rochester, MN 55905 From November 1!)81 through December 1987, 207 patients received whole-abdomen irradiation (WAI) for gynecologic malignancies at the Mayo Clinic. In seven (3%) of these patients, chylous ascites subsequently developed; one additional patient with chylous ascites after WA1 for a gynecologic malignancy was referred to us from another institution. In these eight patients, irradiation was done either adjuvantly (five patients) or as salvage therapy after chemotherapy failure (three patients). Chylous ascites was confirmed by laboratory analysis in six cases and was presumed based on the clinical course in two cases. Mean cumulative radiation doses were 2,925 and 5,122 cGy to the abdomen anld pelvis, respectively, with para-aortic boosts administered in six cases to a mean cumulative dose of approximat’ely 4,200 cGy. The mean time from completion of WA1 to development of ascites was 12 months (range, 6 to 18 months). In six patients, therapy was conservative-observation and diuretics. Two other patients required multiple paracenteses for relief of abdominal distention. Parenteral nutrition was given to two patients who had associated1 radiation enteritis. The ascites resolved in all eight cases at a mean of 18 months (range, 8 to 30 months) after development. At a mean follow-up of 57 months after initial diagnosis and 16 months after resolution of the as’cites, seven patients are without evidence of disease and one patient died of recurrent carcinoma. Distinguishing this clinical entity from recurrent carcinoma is important because of its benign course and its resolution with conservative management. Ovarian cancer, Chylous ascites, Whole-abdomen
radiation therapy.
INTRODUCTION
veloping after whole-abdomen irradiation (WAI) for primary gynecologic malignancies. This report analyzes the time course, natural history, and management of this entity in our experience.
Chylous ascites is a peritoneal effusion, high in fat and protein contents, that results from lymphatic obstruction. The incidence of chylous ascites is difficult to ascertain but recently has been reported to be approximately 1 case per 12,000 admissions at the Massachusetts General Hospital from 1977 to 1982 ( 15). A study by Rovelstad et al. (16) at the Mayo Clinic noted that, in 116 consecutive cases of ascites requiring paracentesis, chylous ascites was found in 10. The causes of chylous ascites have been classified (8) as spontaneous or traumatic. In children most cases are related to congenital lymphatic abnormalities whereas in adults the cause is neoplasm in 80% of cases. Retroperitoneal, mesenteric, or small bowel fibrosis secondary to irradiation or some other influence is an established cause of chylous ascites. Sipes et al. (17) reported two cases of chylous ascites as a sequel of pelvic radiation therapy, and Murray et al. ( 13) described three cases after whole-abdomen radiotherapy for ovarian cancer. We have observed ejght cases of chylous ascites de-
METHODS
Patients and radiation treatments Of the 207 patients who received WA1 for gynecologic malignancy at the Mayo Clinic from 198 1 through 1987, post-irradiation ascites had developed in 7 (3.4%). One additional patient with chylous ascites after adjuvant WA1 for a gynecologic malignancy was referred to us from another institution. The characteristics of these eight patients are presented in Table 1. The mean age was 66 years (range, 47 to 7 1 years); the mean body weight at the initiation of radiation therapy was 68 kg (range, 46 to 88 kg). The mean number of pretreatment operations was 1.5 per patient: two patients had two abdominal explorations, and one patient had
Presented at the Seventeenth Annual Meeting of the Western Association of Gynecologic Oncologists, San Francisco, CA, 1720 May 1989. * Department
of Obstetrizs
AND MATERIALS
+Division of Radiation Oncology. Reprint requests to: Timothy 0. Wilson, M.D., Mayo Clinic, 200 First Street SW, Rochester, MN 55905. Accepted for publication 22 February 1990.
and Gynecology.
435
1. J. Radiation
436
Oncology
0
Biology0 Physics
August
1990. Volume
Patient
Table 1. Characteristics of the eight cases
Radiation
19. Number
Davebpment
Resolution
Length of follow-up
of ascites
of ascites
after rewlution
therapy
Characteristic
Patients, no.
Age, yr 160 >60 Weight, kg 550 >50 Pre-irradiation laparotomy, no. 1 2 3 Para-aortic node dissection Pelvic node dissection
2 6 3 5
Site of
Ovary Ovary
Endometrium
Carcinosarcoma
* After chemotherapy failure. + Received WA1 elsewhere.
z
\NEc!4rnO ANED16nw
.
/+NED~~JM
6-
.
j,+
NED 21 mo
7-
.
//+
NED 24 ma
Z /&ANED29mo I
I
I
18
12
6
0
I
I
I
I
I
I
6
12
la
24
30
36
Fig. 1. Chronologic sequence of events. 0, irradiation; 0, resolution of ascites; A, last follow-up; DOD, dead of disease; NED, no evidence of disease.
Table 2. Characteristics of neoplasms
Serous, grade 3 Serous, grade 4 Serous, grade 2 Mutinous, grade 2 Mutinous, grade 3 Endometrioid, grade 3 Adenosquamous, grade 3
. .
Months
The mean time from completion of WA1 to development of ascites was 12 months (range, 6 to 18 months)
Ovary Ovary Ovary Ovary+ Ovary
3d-
0-a
RESULTS
Histology
-w-‘zmo
5-
three. Pretreatment para-aortic node dissection, performed in three patients, revealed a grossly positive 7-cm mass in one patient; it was completely excised. A pelvic lymphadenectomy was performed in the one patient with endometrial cancer. In the seven cases treated at the Mayo Clinic, WA1 was administered by an open-field technique as described (11). Ovarian malignancies were present in 7 of the 8 cases (Table 2). WA1 was used as salvage treatment after chemotherapy failure in three patients and as primary adjuvant therapy in four cases. The eighth patient received primary adjuvant WA1 for endometrial cancer. Mean cumulative radiation doses, given in 150-cGy daily fractions, were 2,925 and 5,122 cGy to the abdomen and pelvis, respectively, with para-aortic boosts administered in six cases to a mean cumulative dose of 4,200 cGy. There was no therapy in the interval between administration of the radiation and development of chylous ascites.
primary
l-1
2
Stage
Reason for WA1
IV III III I IC
Salvage* Salvage* Salvage* Adjuvant Adjuvant
III
Adjuvant
IIC
Adjuvant
IC
Adjuvant
(Fig. 1). Although some patients were asymptomatic at presentation, most had symptoms including abdominal bloating, dyspepsia, nausea, and vomiting. Examination generally revealed increased abdominal girth, distention, and a fluid wave. CT scans, obtained in five cases, all revealed ascites. Paracentesis was the primary diagnostic modality in seven cases. Peritoneoscopy also was performed in three of these cases. Laparotomy was diagnostic in one patient who had a concurrent small bowel obstruction. The ascitic fluid had a milky appearance in six cases and was serous appearing in two cases. Lipid analysis was performed in six cases (five chylous appearing and one serous) and the results are shown in Table 3. Chylomicrons were the major structural component of the ascitic lipids. In one case of chylous-appearing ascites and one case of serous-appearing ascites, lipid analysis was not performed and the diagnosis of chylous ascites was based on clinical features. Cytologic examination was negative for malignancy in all eight cases. Conservative therapy, consisting of observation and potassium-sparing diuretics, was used as the sole management in six patients. Two other patients required multiple paracenteses for symptomatic relief of abdominal distention. In addition, parenteral nutrition was used for a brief period in two patients as part of the management of associated radiation enteritis. There was no temporal relationship between parenteral nutrition and resolution of the ascites in these two patients. The ascites resolved clinically in all eight cases (CT scan confirmation in five) at a mean of 18 months (range, 8 to 30 months) after its development. At a mean follow-up of 57 months after initial diagnosis and a mean interval of 17 months (range, 3 to 29 months) after resolution of ascites, seven patients Table 3. Lipid analysis of ascitic fluid Test
Mean
Range
Specific gravity Total protein, g/dL Triglyceride, mg/dL Cholesterol, mg/dL
1.022 4.99 1,069 121
1.018-1.025 4.01-6.64 229-2,568 70-203
Chylous ascites after irradiation 0 S. S. LENTZ et al.
are without evidence of disease and one patient has died of recurrent carcinoma. DISCUSSION The diagnosis of ascites generally can be made clinically in conjunction with radiographic techniques if necessary. Paracentesis is confirmatory, allowing diagnostic analysis of the fluid. The fact that the fluid was not chylous in appearance in all of the present cases is not contradictory. Rovelstad et al.(16) found that, among 20 patients with chylous ascites secondary to cancer who had total serum lipids > 350 mg/dL, only 8 had grossly chylous fluid. Thus, exudation of lymph with a relatively high fat content into the peritoneal cavity may occur without necessarily producing fluid with a chylous appearance. The results of the laboratory analyses in .the present series are consistent with those defining chylous ascites in the literature (10). Lymphangiography is of little diagnostic benefit and, in fact, has been associated with a transient exacerbation of the ascites, presumably secondary to further obstruction of compromised lymphatics by the contrast agent-induced inflammatory reaction ( 18). The mean time of 12 months from completion of irradiation to the diagnosis of ascites supports the presumption of radiation injury as the etiologic factor in the development of the chylous ascites. It is postulated, although not specifically documented in the present cases, that radiation-induced fibrosis of the lymphatic vessels within the small bowel and its associated mesentery resulted in occlusion and subsequent extravasation of the chylous fluid. Although a retroperitoneal fibrotic process could be responsible, case reports in the literature support our supposition. Hurst and Edwards (7) described a case of chylous ascites developing after abdominal radiotherapy for recurrent colon carcinoma. At exploration, radiation damage to the small bowel was noted and chyle was leaking from the most severely affected portion. Intraoperative mesenteric lymphangiography revealed lymphatic obstruction in the mesentery. The ch.ylous ascites resolved after resection of the worst-affected segment of small bowel. Similar findings were noted in the three cases reported by Murray and Massey (13) Dilated lymphatics were seen in the root of the small bowel mesentery at laparotomy, suggesting this as the point of obstruction. The possible contribution of a para-aortic boost to the development of chylous ascites could be related to its effect on the base of the small bowel mesentery which is relatively immobile. In addilion, the report by Sipes et al. (17) of chylous ascites after pelvic irradiation also suggests an obstructive lymphatic process within the small bowel and its mesentery. Of interest, in investigating the lymphatic circulation, Blalock et al. (2) noted a lchylous effusion in only 3 of 74 animals undergoing over 250 operative procedures destructive to major lymphatic vessels. At autopsy, numerous lymphaticovenous anastomoses were noted. This
437
suggests that a retroperitoneal lymphatic injury, whether induced surgically or by radiation, rarely leads to chylous ascites. Rather, it is the effect on the intestinal lymphatics and associated lymphaticovenous collaterals that is responsible for the development of chylous ascites. It has been suggested that any process that leads to portal hypertension may predispose to the development of chylous ascites, including radiation-induced hepatitis (6). Even though mild abnormalities in liver function tests were noted in the present cases after irradiation, clinically significant liver dysfunction was not seen. No other consistent predisposing factors were identified in this group of eight cases in comparison with the 207 WA1 cases including a number of pretreatment abdominal operations and para-aortic lymph node dissections. Regarding the present study, the incidence of approximately 3% seen here may be higher than that suggested by case reports and may reflect the relatively large number of patients treated at the Mayo Clinic. In addition, when compared with the large experience at the Princess Margaret Hospital, we used a higher daily fraction size and mean cumulative dose to the entire abdomen (4). In addition, our frequent use of para-aortic boosts may be contributory (6 of the 8 present cases and nearly half of the total patient population receiving WAI). Various therapeutic modalities for chylous ascites have been described based on the above pathophysiology of the disease process (Table 4). Obviously, successful treatment of a neoplastic process that is obstructing efferent lymphatics would be important. Surgical intervention with ligation of traumatized lymphatic channels-for example, as seen after abdominal aortic aneurysm repair and truncal vagotomy-has been described (3, 14). In this situation, instillation of a lipophilic dye into the duodenum immediately before operation has been suggested to help locate the leak (9). Resection of the isolated segment of small intestine responsible for the chylous leakage had been advocated, as described above (7). The presumed diffuse intra-abdominal irradiation effect precluded surgical intervention as a therapeutic option in the cases presented here. Regarding symptomatic management, the use of peritoneovenous shunts, such as the LeVeen shunt, has been described as Table 4. Therapeutic modalities for chylous ascites Primary therapy Treatment for associated malignancy Surgical intervention Repair of leaking lymphatic channel Segmental small bowel resection Symptomatic therapy Peritoneovenous shunt Dietary alterations Medium-chain triglycerides (Tolerex) Parenteral nutrition Multiple paracenteses Diuretics Observation
438
1. J. Radiation Oncology 0 Biology 0 Physics
having generally poor results ( 15). Others have advocated exclusion of long-chain triglycerides from the diet with replacement by dietary supplements (i.e., Tolerex) consisting solely of short- and medium-chain triglycerides (5, 18). This is theoretically sound based on the known physiology of lipid metabolism; however, reported cases have not shown uniform success and the unpalatable nature of this diet is not acceptable to many patients. Total parenteral nutrition (TPN) has been used, thus eliminating enteral long-chain triglycerides ( 1, 12). The benefit from TPN probably is derived from its caloric and protein replacement rather than from the specific exclusion of LCT. The reported variable, and at times transient, response of chylous ascites to TPN generally is observed shortly after its institution. The brief use of parenteral nutrition in the two cases of radiation enteritis in this series was not related temporally to the resolution of the ascites and thus TPN was not considered to be a major therapeutic factor. Diagnostically, paracentesis allows cytologic examination to exclude recurrent neoplasms as well as lipid analysis to confirm the chylous nature of the ascites. In addition, paracentesis provides temporary relief of abdominal distention although it is associated with a small risk of infection and bowel injury. Cytologic examination to exclude recurrent neoplasm should be performed be-
August 1990. Volume 19, Number 2
cause the development of ascites 12 months after therapy suggests recurrence. Finally, medical management using diuretics such as spironolactone has frequently given variable subjective responses; however, it is difficult to evaluate its benefit over observation alone. In the present series, conservative therapy-consisting of observation, multiple paracenteses for symptomatic relief of abdominal distention, and diuretics-was used in all cases. Resolution was achieved in all cases without any obvious associated morbidity. Additionally, the development of chylous ascites was not a harbinger of recurrent disease in any case (based not only on negative cytologic examinations but also on the subsequent clinical course). Although one patient has died of recurrent carcinoma, the ascites had resolved without evidence of disease, as determined by exploratory laparotomy performed for bowel obstruction approximately 1 year prior to the recurrence. Resolution at a mean follow-up time of 18 months also is consistent with the establishment of collateral lymphatic drainage as well as lymphaticovenous anastomoses. The development of ascites is generally considered to be a manifestation of recurrent disease. With the increased use of WA1 as a treatment modality, chylous ascites as a self-limited treatment sequela should be considered in the differential diagnosis.
REFERENCES 1. Asch, M. J.; Sherman, N. J. Management of refractory chylous ascites by total parenteral nutrition. J. Pediatr. 95:260-
262; 1979. 2. Blalock, A.; Robinson, C. S.; Cunningham, R. S.; Gray, M. E. Experimental studies on lymphatic blockage. Arch. Surg. 34:1049-1071; 1937. 3. Bradham, R. R.; Gregorie, H. B.; Wilson, R. Chylous ascites
4. 5.
6.
7.
8. 9.
10.
following resection of an abdominal aortic aneurysm. Ann. Surg. 36:238-240; 1970. Dembo, A. J. Abdominopelvic radiotherapy in ovarian cancer: a IO-year experience. Cancer 55:2285-2290; 1985. Dharman, K.; Temes, S. P.; Wetherell, F. E.; Kendrick, M. J. Chyloperitoneum and chylothorax: a combined rare occurrence after retroperitoneal lymphadenectomy and radiotherapy for testis tumor. J. Urol. 13 1:346-347; 1984. Herz, J.; Shapiro, S. R.; Konrad, P.; Palmer, J. Chylous ascites following retroperitoneal lymphadenectomy: report of 2 cases with guidelines for diagnosis and treatment. Cancer 42:349-352; 1978. Hurst, P. A.; Edwards, J. M. Chylous ascites and obstructive lymphedema of the small bowel following abdominal radiotherapy. Br. J. Surg. 66:780-781; 1979. Kelley, M. L., Jr.; Butt, H. R. Chylous ascites: an analysis of its etiology. Gastroenterology 39:161-170; 1960. Lewis, J. W., Jr.; Storer, E. H. The management ofiatrogenic chylous ascites. Henry Ford Hosp. Med. J. 27: 140-144; 1979. Manten, H.; Barkin, J. S.; Rogers, A. I. Chylous ascites. Postgrad. Med. 7 1:79-84; 1982.
11. Martinez, A.; Schray, M. F.; Howes, A. E.; Bagshaw, M. A. Postoperative radiation therapy for epithelial ovarian cancer: the curative role based on a 24-year experience. J. Clin. Oncol. 3:901-911; 1985. 12. Meinke, A. H., III; Estes, N. C.; Ernst, C. B. Chylous ascites following abdominal aortic aneurysmectomy: management with total parenteral hyperalimentation. Ann. Surg. 190: 63 l-633; 1979. 13. Murray, J. M.; Massey, F. M. Chylous ascites after radiation therapy for ovarian cancer. Obstet. Gynecol. 44:749-751; 1974. 14. Musgrove, J. E. Post-vagotomy Ann. Surg. 175:67-69; 1972.
abdominal
chylous fistula.
15. Press, 0. W.; Press, N. 0.; Kaufman, S. D. Evaluation and management of chylous ascites. Ann. Intern. Med. 96:358364; 1982. 16. Rovelstad, R. A.; Bartholomew, L. G.; Cain, J. C.; McKenzie, B. F.; Soule, E. H. Ascites I. The value of examination of ascitic fluid and blood for lipids and for proteins by electrophoresis. Gastroenterology 34:436-450; 1958. 17. Sipes, S. L.; Newton, M.; Lurain, J. R. Chylous ascites: a sequel of pelvic radiation therapy. Obstet. Gynecol. 66:832835; 1985. 18. Weinstein, L. D.; Scanlon, G. T.; Hersh, T. Chylous ascites management with medium-chain triglycerides and exacerbation by lymphangiography. Am. J. Digest. Dis. 14:500509; 1969.