LB128 - CLINICAL CHARACTERISTICS AND EVOLUTION OF A COHORT OF PATIENTS WITH CHYLOUS FISTULAS ATTENDED IN A REFERRAL HOSPITAL UNDER ROUTINE CLINICAL PRACTICE

LB128

CLINICAL CHARACTERISTICS AND EVOLUTION OF A COHORT OF PATIENTS WITH CHYLOUS FISTULAS ATTENDED IN A REFERRAL HOSPITAL UNDER ROUTINE CLINICAL PRACTICE

A. Jiménez-Sánchez^1,*, S. Torres-Degayón¹, I. González-Navarro¹, L. González-Gracia¹, A. J. Martínez-Ortega¹, J. L. Pereira-Cunill¹, P. P. García-Luna¹

¹Unidad de Endocrinología y Nutrición, Hospital Universitario Virgen del Rocío (Servicio Andaluz de Salud), Sevilla, Spain

 

Rationale: Chylous fistulas are an infrequent complication, frequently related to complex surgical procedures. This condition can be optimized using nutritional treatment in the context of a multi-disciplinary approach. 

Our goal was to analyze the clinical presentation and evolution of a cohort of patients with chylous fistulas that were attended in a real-life setting.

Methods:         - Consecutive sampling was used. All incident cases of chylous fistulas with interconsultation to our Clinical Nutrition Unit and completed follow-up from June 202 to September 2024 were included for analysis.

        - Variables: data was retrospectively collected using an electronic clinical history. Demographics and clinical data were gathered. The following analytes were also monitored during follow-up: lymphocytes (n and %), total proteins (g/dL), Na and K (mEq/L), Ca, P, and Mg (mg/dL).

        - Statistics: numerical variables were synthesized as median (IQR), and frequencies expressed as count (%). We compared the differential clinical characteristics of the participants depending on fistula location (cervical, thoracic, or abdominal) and whether the surgeries were non-oncological or oncological, as well as the evolution of laboratory parameters over time. We used Kruskal-Wallis tests and post-hoc p-adjusted pairwise comparisons using Wilcoxon tests if the corresponding variables were numeric; otherwise, we used Fisher's test.  Non-significant p-values are reported as "ns". 

Results:         - n = 20 patients were analyzed, of whom 11/20 (55%) were men. Age was 67 (21) years.

        - Cancers were the main subjacent disease: pancreas 4/20 (20%), lymphoma 3/20 (15%), neuroendocrine tumors 2/20 (10%), and sarcoma 2/20 (10%), among others. The surgical approach was diverse in this group, with a majority of laparoscopy (6/13, 46%).

        - There were 4/20 (20%) non-malignant cases: liver transplant, toxic megacolon, hysterocele, and mediastinal schwannoma. The main surgical approach in this group was laparotomy (3/4, 75%).

        - Fistula location was mainly abdominal 13/20 (65%), followed by thoracic 4/20 (20%), and cervical 3/20 (15%).

        - Time to diagnosis was 7(5) days, with no differences regarding fistula location or oncological surgery (ns).

        - Macroscopical drainage aspect was used for diagnosis in 100% in cervical and 10/13 (77%) of abdominal fistulas, while 1/4 (25%) of thoracic fistulas had this feature. The rest were diagnosed via centesis (p = 0.078).

        - Peak output (mL/day) varied by location: 70(15) in cervical, 1,200 (200) in thoracic, and 800 (1,150) in abdominal (p = 0.021) fistulas. Cervical fistulas output was lower than that of abdominal ones (p = 0.045) and tended to be lower than thoracic fistulas (p = 0.083). Non-oncological surgeries displayed higher baseline output [5,438 (9,444) vs 250 (400) mL/day; p = 0.009] and peak output [6,000 (8,531) vs 350 (700); p = 0.067]. The highest peak output in the sample (11,000 mL/day) was due to hepatic trasplantation, and the second due to toxic megacolon (10,000 mL/day).

        - Time to closure (days) tended to vary by location: 135(188) in thoracic, 22(45) in cervical, and 13(5) in abdominal fistulas (p = 0.063), although no differences were found in post-hoc pairwise comparisons. No differences were found regarding non-oncological vs oncological surgeries (ns).

        - Baseline BMI was 24.2 (5.1) kg/m². Malnutrition (by Subjective Global Assessment) was present in 7/20 (35%) cases, with 3/7 (42.8%) severe, 3/7 (42.8%) moderate and 1/7 (14.3%) mild cases, with no differences regarding fistula location or non-oncological vs oncological surgeries (ns).

        - Enteral tube feeding was used in 1/3 (33%) participants with cervical fistulas, while the rest of the sample received oral nutritional supplements (ns). We found no differences regarding non-oncological vs oncological surgeries (ns).

        - Patients with cervical fistulas did not need parenteral nutrition, while some abdominal (4/13, 31%) and thoracic (2/4, 50%) cases did (ns). Patients with non-oncological surgery tended to have more parenteral nutricion than oncological surgeries: 3/4 (75%) vs 2/13 (15%); p = 0.053.

        - Octreotide was prescribed in all patients with thoracic fistulas, as well as in some cervical (1/3, 33%) and abdominal (10/13, 77%) cases, tending to reach signficance (p = 0.111). We found no differences regarding non-oncological vs oncological surgeries (ns).

        - Lymphocytes (p = 0.01), total proteins (p <0.001) and K (p = 0.015) varied during follow-up in the whole sample, while the other laboratory parameters did not (ns).

        - Regarding complications, hydroelectrolytic alterations appeared in 10/20 (50%) cases (ns), infections in 6/20 (30%) participants (ns), and exitus in 2/20 (10%) patients (ns). Non-oncological surgeries displayed a higher proportion of hydroelectrolitic alterations: 4/4 (100%) vs 4/13 (31%); p = 0.029. No differences in infections or exitus were found regarding non-oncological vs oncological surgeries.

        - All survivors achieved fistula closure at hospital discharge.

Conclusion:         - Chylous fistulas were clinically heterogeneous.

        - Output was lower in cervical fistulas.

        - Non-malignant surgical cases displayed more severe fistulas, with a higher output, more hydroelectrolytical alterations, and a tendency to more parenteral nutrition. This may be attributed to a higher prevalence of laparotomy in this group.

        - Chylous fistulas in surviving patients responded well to nutritional treatment.

Disclosure of Interest: None declared