- Email: [email protected]
Use of Monogen for pediatric postoperative chylothorax
Barbara E. Cormack, NZRD, Nigel J. Wilson, FRACP, Kirsten Finucane, FRACS, and Teena M. West, MS Nutrition Services, Pediatric Cardiology Department, Cardiothoracic Surgical Unit, and Biostatistics Department, Green Lane Hospital, Auckland, New Zealand
Background. Postoperative chylothorax is a potentially serious complication of pediatric cardiac surgery. The purpose of this study was to report the use of Monogen enteral formula for the management of pediatric postoperative chylothorax. Methods. A retrospective, single-institution 2-year study of all patients with a diagnosis of postoperative chylothorax was conducted. Chylothorax was diagnosed in 25 patients after a total of 535 cardiac operations in children younger than 10 years, for an incidence of 4.7%. Eighteen patients had been given Monogen, an enteral low long-chain triglyceride formula, as initial treatment. Six had been given total parenteral nutrition. The following variables were related to outcome and response to Monogen: age, sex, weight, underlying condition, type of surgery, interval between surgery and chylothorax diag-
nosis, duration and daily volume of chyle leak, central venous pressure, residual lesions, and weight loss. Results. Enteral feeding with Monogen was successful for 14 of 18 patients with a response to treatment evident by the end of the third day. No variables predicted which patients would respond to Monogen. Body weight was maintained or increased in 14 of the 17 surviving patients taking Monogen. A return to normal diet at 4 ⴞ 1 weeks from the day of pleural drain removal did not result in recurrent chylothorax. Conclusions. A trial of Monogen is recommended as initial treatment for postoperative chylothorax unless enteral feeding is contraindicated.
P
Material and Methods
ostoperative chylothorax is a potentially serious complication of pediatric cardiac surgery. The general consensus is that the management of pediatric postoperative chylothorax should include chest tube drainage and nutritional support, but little agreement has been reached about the type of nutritional support, the need for gut rest, or the role of pharmacologic agents or surgical procedures. Total parenteral nutrition (TPN) and medium-chain triglyceride (MCT) diets have variable success as treatments for chylothorax [1– 6]. No randomized trials have been conducted to compare the efficacy of these different forms of treatment. From May 1999 we began using Monogen (Scientific Hospital Supplies, Liverpool, England) instead of Portagen powder as enteral nutrition support for chylothorax. Monogen is a nutritionally complete, low-fat, whole whey protein, powdered feed containing 93% of fat as MCT. Compared with Portagen, Monogen has a lower level of long-chain triglycerides (LCT; 1.4 g/L) and a higher energy level of 74 kcal/100 mL; thus Monogen was thought to be a better choice (Table 1). The purpose of this study was to audit our use of Monogen for postoperative chylothorax and to formulate treatment guidelines in light of these results.
Accepted for publication July 1, 2003. Address reprint requests to Ms Cormack, Nutrition Services, Green Lane Hospital, Greenlane West, Auckland 1030, New Zealand; e-mail: [email protected].
© 2004 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2004;77:301–5) © 2004 by The Society of Thoracic Surgeons
From June 1999 Monogen was used as the enteral feed for postoperative chylothorax in our institution. A retrospective study of all pediatric patients at Green Lane Hospital treated for postoperative chylothorax for the next 2 years was conducted. Patients were identified by the hospital diagnostic coding index and dietitians’ records. During the 2-year period postoperative chylothorax was diagnosed in 25 cases after a total of 535 cardiac operations in children younger than 10 years, an incidence of 4.7% in this age group. Data on age, sex, weight, underlying condition, type of surgery, interval between surgery and chylothorax diagnosis, duration (defined as days from chylothorax diagnosis to removal of pleural drains) daily volume of chyle leak, average central venous pressure on day of diagnosis, residual lesions, reoperation, thoracic duct ligation, weight loss, and outcome were collected from medical and dietitians’ records. Only postoperative chylothoraces were included. Surgical operations were arbitrarily categorized as neonatal surgery (neonatal arch obstruction, Norwood or arterial switch) (n ⫽ 8), tetralogy of Fallot or pulmonary artery surgery (n ⫽ 6), bidirectional Glenn or Fontan procedures (n ⫽ 6), and left-to-right shunt complete repairs (n ⫽ 5). A residual cardiac lesion was diagnosed if the patients had moderate or severe left ventricular impairment, significant residual ventricular septal defect, moderate or severe atrioventricular valve regurgitation, or significant pulmonary artery stenosis. The first day of treatment for chylothorax was defined as day 1. 0003-4975/04/$30.00 doi:10.1016/S0003-4975(03)01189-5
GENERAL THORACIC
Use of Monogen for Pediatric Postoperative Chylothorax
302
CORMACK ET AL PEDIATRIC POSTOPERATIVE CHYLOTHORAX
Ann Thorac Surg 2004;77:301–5
Table 1. Comparison of Monogen and Portagen Composition Per Liter GENERAL THORACIC
Energy (kcal) Fat (g) % MCT LCT per Litre (g) Protein Osmolality at standard dilution (mosm/kg) Essential fatty acids provided by
Monogen
Portagen
740 20 93 1.4
670 30 87 3.9
Whey protein concentrate and amino acids 280
Sodium caseinate
Walnut oil
Corn oil
LCT ⫽ long-chain triglyceride;
350
MCT ⫽ medium-chain triglyceride.
The diagnosis of chylothorax was based on the presence of a persistent sterile effusion and confirmed by examination of the pleural fluid. Regardless of volume, a noninfectious milky appearance in orally fed patients with persistent chylous pleural effusion was considered diagnostic. A triglyceride level had been measured in 16 of 25 cases, and the concentration was higher than 1.1 mmol/L (⬎ 110 mg/dL). Patients with the largest drain volumes tended not to have their triglyceride levels measured, as the diagnosis was intuitive. Baseline characteristics were compared between the group who responded to Monogen and Monogen nonresponders. Response to Monogen was defined as recovery from the chylothorax with no further form of treatment (parenteral nutrition or surgery) required. The characteristics of patients who responded to either Monogen or TPN, defined as nutritional treatment, were also compared with those who did not respond to nutritional treatment. Response to nutritional treatment was defined as recovery from the chylothorax with no further surgery required. Summary statistics have been presented as median and range. Baseline characteristics were compared using a Mann-Whitney U test, which compares the medians of two samples and is the nonparametric equivalent of a two-sample unpaired t test. The Fisher exact test was also used where appropriate. A nonparametric test has been used because of the small, unbalanced sample sizes and uncertainty about the distribution of the population from which the samples were drawn.
Results Twenty-five cases were identified, of whom 23 patients survived. Nineteen cases had enteral nutrition as their initial form of treatment. Of these, 18 had Monogen; 1 older child who presented 24 days postoperatively had 4 days on Monogen and a minimal LCT fat diet. Due to poor nutritional status rather than failure to respond to the diet this patient was changed to TPN. He was excluded from further analysis. Six patients, who were in the intensive care unit (ICU) at the time of diagnosis of chylothorax, received TPN as their initial form of nutritional treatment.
Fourteen of the 18 cases (78%) resolved on Monogen. Table 2 shows the baseline characteristics and outcome of the Monogen group. In the Monogen group, a substantial decrease of 17.3 to 6.1 mL/kg per day in median pleural drainage volume signaled a response to treatment by the end of the third day (Fig 1). Overall duration of the chylothorax from day of diagnosis for the Monogen group varied from 2 to 36 days, with a median of 4.5 days of drainage. The length of stay from surgery to discharge ranged from 8 to 74 days. The average surgical length of stay for the unit is 8 days. Of the 4 patients whose condition did not resolve, 2 had pleural drainage volumes that were still increasing after 9 days on Monogen. The other 2 patients experienced initial resolution on Monogen and were discharged, but were readmitted with reaccumulation of chylous effusion while still on Monogen after 1 and 2 weeks, respectively. These 4 patients were changed to TPN. Two recovered on TPN. One of the patients who had reaccumulated chyle while at home on Monogen required thoracic duct ligation. One patient with severely impaired ventricular function after neonatal arterial switch surgery developed severe mitral regurgitation requiring acute valve replacement. He died of multiorgan failure, although the chylothorax had resolved. Six patients had TPN as their initial treatment because they were unstable neonates or infants with high pleural drain volumes in the ICU rather than in the cardiac ward. Table 2 shows the baseline characteristics and outcome of the TPN group. Two recovered on TPN after 8 and 9 days, respectively, without further intervention. Two of the 4 who did not resolve underwent thoracic duct ligation. One had revisional surgery for a residual ventricular septal defect. The sixth patient Table 2. Baseline Characteristics and Outcome of the Monogen and TPN Groups
Patient Characteristic Age, months Sex, male, n (%) Bilateral chylothoraces, n (%) Drain volume on day of diagnosis (mL/kg/day) Average CVP on day of diagnosis (mm Hg)a Triglycerides (mmol/L)b Chyle leak diagnosed after surgery (days) Residual lesions, n (%) Duration of drainage from diagnosis (days) Peak 24-hour pleural drain (mL/kg per day) Time from surgery to discharge (days) a CVP monitored in 16 patients. patients.
Monogen Group (n ⫽ 18)
TPN (n ⫽ 6)
5.5 (0.3, 72) 14 (74%) 5 (26%)
1.7 (0.3, 11.6) 4 (67%) 2 (33%)
17.3 (4.6, 32)
29.5 (18.8, 32.1)
10 (5, 16) 4.6 (2.5, 57.1) 5 (1, 12)
9 (5, 12) 6.3 (2.3, 19.8) 6 (4, 11)
8 (44%) 4.5 (3, 9)
5 (83%) 18.5 (16, 28)
31 (14, 73)
73 (42, 105)
18 (8, 74)
39 (19, 57)
b
Triglycerides measured in 11
Values presented as median (interquartile range) unless otherwise noted. CVP ⫽ central venous pressure;
TPN ⫽ total parenteral nutrition.
303
CORMACK ET AL PEDIATRIC POSTOPERATIVE CHYLOTHORAX
pressure, peak 24-hour volumes, triglyceride concentrations, or days from surgery to diagnosis. Three of the 4 nonresponders had residual lesions, but this difference was not statistically significant (p ⫽ 0.5865) All patients who responded to Monogen had decreased 24-hour drain volumes to less than 20 mL/kg by the end of day 4.
Nutritional Treatment Responders Versus Nonresponders A comparison of the 16 nutritional treatment (Monogen or TPN) responders with the 8 nonresponders showed no significant difference in age, weight, sex, bilateral pleural effusion, central venous pressure, peak 24-hour volumes, triglycerides, residual lesions, or days from surgery to diagnosis. Fig 1. Median volume of pleural drainage: Monogen (n ⫽ 18) group.
Comment Monogen
died of irreversible pulmonary hypertension following repair of a Taussig Bing anomaly with hypoplastic right lung. Overall, 18 of the original 24 cases (75%) recovered after treatment with Monogen or TPN without need for further surgical intervention. Fourteen of the surviving 17 Monogen patients gained weight from the day of their surgery to hospital discharge. The remaining 3 cases were discharged weighing less than their weight at surgery, although their weight loss was less than 3%. Most of the group went home with Monogen treatment. The interval from the last pleural drain removal to the change to normal formula/diet varied from 1 to 15 weeks. Fifteen patients (68%) stayed on Monogen for 4 ⫾ 1 weeks, and none of the patients reaccumulated a chylothorax after changing back to their normal diet.
Monogen is an MCT-based enteral formula designed for infants and children with lipid and lymphatic disorders. The MCT in both Portagen and Monogen is provided by fractionated coconut oil, but the fatty acid profile of Monogen is unique because it contains 93% MCT with only 7% LCT provided by walnut oil. This LCT level is sufficient to meet essential fatty acid requirements. In contrast, Portagen has 14% of total fat as LCT (4.2 g/L). The osmolality of Monogen is substantially lower than Portagen and most other elemental or fat-free formulas, a difference that improves gastrointestinal tolerance of the feed. Monogen also has a substantially higher energy level than alternative feeds. This point is important because patients with chylothorax have high energy requirements due to increased metabolic demand from the combination of chyle loss and hypermetabolism associated with surgery. We have found that Monogen is an effective form of treatment for postoperative chylothorax in children. Chylothoraces had resolved by 3.5 days in 78% of patients. This rapid response with Monogen compares well with
Monogen Responders Versus Nonresponders A comparison of the 14 Monogen responders with the 4 nonresponders showed no significant difference in age, weight, sex, bilateral pleural effusion, central venous
Table 3. Reported Chylothorax Incidence and Treatment Outcomes in Children Resolved on Enteral Nutrition Cases (n)
Incidence (%)
Survival (%)
Puntis Allen7
15 18
a
1 0.9
93 94
Nguyen3
24
1.9
79
Bond5 Beghetti12
25 51
1.5 2.5
88 90
Current study
25
4.7
92
Study 9
a
Treatment
%
Mean Duration (days)
Low-fat diet ⫹ MCT Low-fat diet (Portagen) Low-fat diet ⫹ TPN Low-fat diet ⫹ MCT (Portagen or Vivonex) or TPN only TPN only Low-fat diet for 1 week then TPN Monogen
73 67
12 11.3
73 39
12 7
66
27
84
15.7
73 80
11.9 24.7
75
Median 5.0
One percent of patent ductus arteriosis procedures.
MCT ⫽ medium-chain triglyceride;
Resolved on Enteral or TPN
TPN ⫽ total parenteral nutrition.
Not stated
Not stated
78
Median 3.5
%
Mean Duration (days)
GENERAL THORACIC
Ann Thorac Surg 2004;77:301–5
304
CORMACK ET AL PEDIATRIC POSTOPERATIVE CHYLOTHORAX
Ann Thorac Surg 2004;77:301–5
GENERAL THORACIC Fig 2. Algorithm of management of chylothorax. (LCT ⫽ long-chain triglyceride.)
other forms of enteral feeding such as Portagen [3, 6, 7], Vivonex [3], and “fat-free formula” [8]. A positive response with reduction of chyle drainage occurred by the end of the third day. Based on our results, in the absence of contraindications to enteral feeding, we believe an initial trial of enteral feeding using Monogen is appropriate. This study also shows that Monogen provides effective nutritional treatment for chylothorax. Body weight in our Monogen group was well maintained with 82% of our patients discharged weighing more than their preoperative weight. Weight loss was minimal in the other 18% (n
⫽ 3), less than 3% of total body weight. More significant weight loss has been seen with other patients with chylothorax when treated with other enteral formulas [7, 9]. Puntis and colleagues [9] reported 8% weight loss in a third of patients. Allen and colleagues [7] reported a higher than 10% weight loss in 22% of their cohort treated with Portagen. We have not found other published reports on the use of Monogen for management of chylothorax. Previous studies have reported use of Portagen [3, 6, 7]. Table 3 summarizes studies using a variety of enteral and parenteral treatments. Several research groups have agreed
that an enteral minimal LCT fat diet is as effective as TPN in controlling lymph leakage in appropriately selected patients [3, 7, 10].
Parenteral Nutrition During this study TPN had been used as initial treatment in 6 patients. The unequal sample size and baseline characteristics of this TPN group and the Monogen group made statistical comparison with the Monogen group potentially misleading. However, the frequent and rapid response with Monogen compares well with this small TPN experience and with others using TPN [5]. Bond and associates [5] found the average duration of drainage was 11.9 days and resolution occurred in 17 of 19 patients, but only after 20 days of total gut rest. The rationale in our ICU for using TPN as the initial treatment was that enteral feeding may be poorly absorbed in the sick neonate or infant, resulting in a poorer nutritional status and the risk of precipitating necrotizing enterocolitis. In a few cases severe fluid restriction may necessitate the use of concentrated TPN to meet nutritional requirements. However, TPN has disadvantages, including gut stasis, catheter-related sepsis, and cost. Our success with Monogen has led us to adjust our initial policy and we now intend to use the formula in ICU patients unless enteral feeding is completely contraindicated. If the pleural drainage is not decreasing by day 5 then TPN should be considered, but investigation for residual cardiac lesions is warranted and early reoperation should be undertaken where appropriate.
Timing of Return to Normal Diet It is generally recommended that patients stay on the MCT formula or minimal LCT fat diet for 3 to 6 weeks, but there is no clear consensus about this strategy or whether to begin timing the treatment from the date of diagnosis, pleural drain removal, finishing TPN, or discharge from hospital [3, 10, 11]. Staying on this restrictive diet longer than necessary is expensive and may mean breast-feeding is not reestablished. This study provides some data about the timing of return to normal diet. None of the patients reaccumulated chyle after they were switched to a normal diet within 6 weeks, nor did the 6 patients who received Monogen for less than 4 weeks. This result confirms 6 weeks on the low LCT diet from the day of the last pleural drain removal is long enough to avoid recurrence. We currently recommend 4 weeks for the duration of Monogen or low LCT diet. An algorithm for the treatment of pediatric postoperative chylothorax based on our findings and review of the literature has now been developed (Fig 2).
Incidence of Chylothorax and Survival The chylothorax incidence rate of 4.7% for the 2-year period in this study is higher than in previous reports. Table 3 shows the reported incidence from five reports published since 1987. Comparison of studies is confounded by varying definitions for the denominator of “total cardiac operations or procedures.” Most authors did not precisely define “total cardiac procedures.” We
CORMACK ET AL PEDIATRIC POSTOPERATIVE CHYLOTHORAX
305
have taken cardiothoracic surgery procedures in patients 10 years of age or younger as the denominator. Some centers may make the diagnosis of chylothorax only when a significant volume or duration is reached. In our study, the mean interval from surgery to diagnosis was 5.9 days in the Monogen group (range 1 to 21 days). This interval is shorter than Le Coultre’s reported mean of 9.8 days and Beghetti’s mean of 7.3 days for those who recovered quickly and 14 days for those with longstanding chylous losses [10, 12]. Earlier recognition and treatment in our group may also help to explain the apparently higher incidence of chylothorax in our unit. The reported incidence appears to increase as more complex surgery is performed [12]. Additionally, a recent trend has been to start feeding patients earlier after surgery, a strategy that may improve nutritional status but that has an unknown influence on the development of chylothorax. With 92% survival, our group compares well with other published survival rates as summarized in Table 3.
Conclusions Monogen is an effective high-energy enteral nutrition treatment for postoperative chylothorax in children. Response is usually rapid within 5 days and body weight is maintained. Treatment can be discontinued after 4 weeks without a recurrence of chylothorax.
References 1. Ramos W, Faintuch J. Nutritional management of thoracic duct fistulas. A comparative study of parenteral versus enteral nutrition. J Parenter Enteral Nutr 1986;10:519 –21. 2. Shaw V, Lawson M. The cardiothoracic system. In: Shaw V, Lawson M, eds. Clinical pediatric dietetics. London: Blackwell Scientific, 2001:185–92. 3. Nguyen DM, Shum-Tim D, Dobell AR, Tchervenkov CI. The management of chylothorax chylopericardium following pediatric cardiac surgery: a 10-year experience. J Card Surg 1995;10:302–8. 4. Marts BC, Naunheim KS, Fiore AC, et al. Conservative versus surgical management of chylothorax. Am J Surg 1992;164:532–4. 5. Bond SJ, Guzzetta PC, Snyder ML, Randolph JG. Management of pediatric postoperative chylothorax. Ann Thorac Surg 1993;56:469 –73. 6. Koloske AM, Martin LW, Schubert WK. Management of chylothorax in children by thoracentesis and medium-chain triglyceride feedings. J Pediatr Surg 1974;9:365–71. 7. Allen EM, van Heeckeren DW, Spector ML, Blumer JL. Management of nutritional and infectious complications of post-operative chylothorax in children. J Pediatr Surg 1991; 26:1169 –74. 8. Buttiker V, Fanconi S, Burger R. Chylothorax in children. Chest 1999;116:682–7. 9. Puntis JW, Roberts KD, Handy D. How should chylothorax be managed? Arch Dis Child 1987;62:593–6. 10. Le Coultre C, Oberhansli I, Mossaz A, Bugmann P, Faidutti B, Belli DC. Postoperative chylothorax in children: difference between vascular and traumatic origin. J Pediatr Surg 1991; 26:519 –23. 11. Bonner G, Warren J. A review of the nutritional management of chyle leakage in adults. J Hum Nutr Diet 1998;11: 105–14. 12. Beghetti M, La Scala G, Belli D, Bugmann P, Kalangos A, Le Coultre C. Etiology and management of pediatric chylothorax. J Pediatr 2000;136:653–8.
GENERAL THORACIC
Ann Thorac Surg 2004;77:301–5
Background. Postoperative chylothorax is a potentially serious complication of pediatric cardiac surgery. The purpose of this study was to report the use of Monogen enteral formula for the management of pediatric postoperative chylothorax. Methods. A retrospective, single-institution 2-year study of all patients with a diagnosis of postoperative chylothorax was conducted. Chylothorax was diagnosed in 25 patients after a total of 535 cardiac operations in children younger than 10 years, for an incidence of 4.7%. Eighteen patients had been given Monogen, an enteral low long-chain triglyceride formula, as initial treatment. Six had been given total parenteral nutrition. The following variables were related to outcome and response to Monogen: age, sex, weight, underlying condition, type of surgery, interval between surgery and chylothorax diag-
nosis, duration and daily volume of chyle leak, central venous pressure, residual lesions, and weight loss. Results. Enteral feeding with Monogen was successful for 14 of 18 patients with a response to treatment evident by the end of the third day. No variables predicted which patients would respond to Monogen. Body weight was maintained or increased in 14 of the 17 surviving patients taking Monogen. A return to normal diet at 4 ⴞ 1 weeks from the day of pleural drain removal did not result in recurrent chylothorax. Conclusions. A trial of Monogen is recommended as initial treatment for postoperative chylothorax unless enteral feeding is contraindicated.
P
Material and Methods
ostoperative chylothorax is a potentially serious complication of pediatric cardiac surgery. The general consensus is that the management of pediatric postoperative chylothorax should include chest tube drainage and nutritional support, but little agreement has been reached about the type of nutritional support, the need for gut rest, or the role of pharmacologic agents or surgical procedures. Total parenteral nutrition (TPN) and medium-chain triglyceride (MCT) diets have variable success as treatments for chylothorax [1– 6]. No randomized trials have been conducted to compare the efficacy of these different forms of treatment. From May 1999 we began using Monogen (Scientific Hospital Supplies, Liverpool, England) instead of Portagen powder as enteral nutrition support for chylothorax. Monogen is a nutritionally complete, low-fat, whole whey protein, powdered feed containing 93% of fat as MCT. Compared with Portagen, Monogen has a lower level of long-chain triglycerides (LCT; 1.4 g/L) and a higher energy level of 74 kcal/100 mL; thus Monogen was thought to be a better choice (Table 1). The purpose of this study was to audit our use of Monogen for postoperative chylothorax and to formulate treatment guidelines in light of these results.
Accepted for publication July 1, 2003. Address reprint requests to Ms Cormack, Nutrition Services, Green Lane Hospital, Greenlane West, Auckland 1030, New Zealand; e-mail: [email protected].
© 2004 by The Society of Thoracic Surgeons Published by Elsevier Inc
(Ann Thorac Surg 2004;77:301–5) © 2004 by The Society of Thoracic Surgeons
From June 1999 Monogen was used as the enteral feed for postoperative chylothorax in our institution. A retrospective study of all pediatric patients at Green Lane Hospital treated for postoperative chylothorax for the next 2 years was conducted. Patients were identified by the hospital diagnostic coding index and dietitians’ records. During the 2-year period postoperative chylothorax was diagnosed in 25 cases after a total of 535 cardiac operations in children younger than 10 years, an incidence of 4.7% in this age group. Data on age, sex, weight, underlying condition, type of surgery, interval between surgery and chylothorax diagnosis, duration (defined as days from chylothorax diagnosis to removal of pleural drains) daily volume of chyle leak, average central venous pressure on day of diagnosis, residual lesions, reoperation, thoracic duct ligation, weight loss, and outcome were collected from medical and dietitians’ records. Only postoperative chylothoraces were included. Surgical operations were arbitrarily categorized as neonatal surgery (neonatal arch obstruction, Norwood or arterial switch) (n ⫽ 8), tetralogy of Fallot or pulmonary artery surgery (n ⫽ 6), bidirectional Glenn or Fontan procedures (n ⫽ 6), and left-to-right shunt complete repairs (n ⫽ 5). A residual cardiac lesion was diagnosed if the patients had moderate or severe left ventricular impairment, significant residual ventricular septal defect, moderate or severe atrioventricular valve regurgitation, or significant pulmonary artery stenosis. The first day of treatment for chylothorax was defined as day 1. 0003-4975/04/$30.00 doi:10.1016/S0003-4975(03)01189-5
GENERAL THORACIC
Use of Monogen for Pediatric Postoperative Chylothorax
302
CORMACK ET AL PEDIATRIC POSTOPERATIVE CHYLOTHORAX
Ann Thorac Surg 2004;77:301–5
Table 1. Comparison of Monogen and Portagen Composition Per Liter GENERAL THORACIC
Energy (kcal) Fat (g) % MCT LCT per Litre (g) Protein Osmolality at standard dilution (mosm/kg) Essential fatty acids provided by
Monogen
Portagen
740 20 93 1.4
670 30 87 3.9
Whey protein concentrate and amino acids 280
Sodium caseinate
Walnut oil
Corn oil
LCT ⫽ long-chain triglyceride;
350
MCT ⫽ medium-chain triglyceride.
The diagnosis of chylothorax was based on the presence of a persistent sterile effusion and confirmed by examination of the pleural fluid. Regardless of volume, a noninfectious milky appearance in orally fed patients with persistent chylous pleural effusion was considered diagnostic. A triglyceride level had been measured in 16 of 25 cases, and the concentration was higher than 1.1 mmol/L (⬎ 110 mg/dL). Patients with the largest drain volumes tended not to have their triglyceride levels measured, as the diagnosis was intuitive. Baseline characteristics were compared between the group who responded to Monogen and Monogen nonresponders. Response to Monogen was defined as recovery from the chylothorax with no further form of treatment (parenteral nutrition or surgery) required. The characteristics of patients who responded to either Monogen or TPN, defined as nutritional treatment, were also compared with those who did not respond to nutritional treatment. Response to nutritional treatment was defined as recovery from the chylothorax with no further surgery required. Summary statistics have been presented as median and range. Baseline characteristics were compared using a Mann-Whitney U test, which compares the medians of two samples and is the nonparametric equivalent of a two-sample unpaired t test. The Fisher exact test was also used where appropriate. A nonparametric test has been used because of the small, unbalanced sample sizes and uncertainty about the distribution of the population from which the samples were drawn.
Results Twenty-five cases were identified, of whom 23 patients survived. Nineteen cases had enteral nutrition as their initial form of treatment. Of these, 18 had Monogen; 1 older child who presented 24 days postoperatively had 4 days on Monogen and a minimal LCT fat diet. Due to poor nutritional status rather than failure to respond to the diet this patient was changed to TPN. He was excluded from further analysis. Six patients, who were in the intensive care unit (ICU) at the time of diagnosis of chylothorax, received TPN as their initial form of nutritional treatment.
Fourteen of the 18 cases (78%) resolved on Monogen. Table 2 shows the baseline characteristics and outcome of the Monogen group. In the Monogen group, a substantial decrease of 17.3 to 6.1 mL/kg per day in median pleural drainage volume signaled a response to treatment by the end of the third day (Fig 1). Overall duration of the chylothorax from day of diagnosis for the Monogen group varied from 2 to 36 days, with a median of 4.5 days of drainage. The length of stay from surgery to discharge ranged from 8 to 74 days. The average surgical length of stay for the unit is 8 days. Of the 4 patients whose condition did not resolve, 2 had pleural drainage volumes that were still increasing after 9 days on Monogen. The other 2 patients experienced initial resolution on Monogen and were discharged, but were readmitted with reaccumulation of chylous effusion while still on Monogen after 1 and 2 weeks, respectively. These 4 patients were changed to TPN. Two recovered on TPN. One of the patients who had reaccumulated chyle while at home on Monogen required thoracic duct ligation. One patient with severely impaired ventricular function after neonatal arterial switch surgery developed severe mitral regurgitation requiring acute valve replacement. He died of multiorgan failure, although the chylothorax had resolved. Six patients had TPN as their initial treatment because they were unstable neonates or infants with high pleural drain volumes in the ICU rather than in the cardiac ward. Table 2 shows the baseline characteristics and outcome of the TPN group. Two recovered on TPN after 8 and 9 days, respectively, without further intervention. Two of the 4 who did not resolve underwent thoracic duct ligation. One had revisional surgery for a residual ventricular septal defect. The sixth patient Table 2. Baseline Characteristics and Outcome of the Monogen and TPN Groups
Patient Characteristic Age, months Sex, male, n (%) Bilateral chylothoraces, n (%) Drain volume on day of diagnosis (mL/kg/day) Average CVP on day of diagnosis (mm Hg)a Triglycerides (mmol/L)b Chyle leak diagnosed after surgery (days) Residual lesions, n (%) Duration of drainage from diagnosis (days) Peak 24-hour pleural drain (mL/kg per day) Time from surgery to discharge (days) a CVP monitored in 16 patients. patients.
Monogen Group (n ⫽ 18)
TPN (n ⫽ 6)
5.5 (0.3, 72) 14 (74%) 5 (26%)
1.7 (0.3, 11.6) 4 (67%) 2 (33%)
17.3 (4.6, 32)
29.5 (18.8, 32.1)
10 (5, 16) 4.6 (2.5, 57.1) 5 (1, 12)
9 (5, 12) 6.3 (2.3, 19.8) 6 (4, 11)
8 (44%) 4.5 (3, 9)
5 (83%) 18.5 (16, 28)
31 (14, 73)
73 (42, 105)
18 (8, 74)
39 (19, 57)
b
Triglycerides measured in 11
Values presented as median (interquartile range) unless otherwise noted. CVP ⫽ central venous pressure;
TPN ⫽ total parenteral nutrition.
303
CORMACK ET AL PEDIATRIC POSTOPERATIVE CHYLOTHORAX
pressure, peak 24-hour volumes, triglyceride concentrations, or days from surgery to diagnosis. Three of the 4 nonresponders had residual lesions, but this difference was not statistically significant (p ⫽ 0.5865) All patients who responded to Monogen had decreased 24-hour drain volumes to less than 20 mL/kg by the end of day 4.
Nutritional Treatment Responders Versus Nonresponders A comparison of the 16 nutritional treatment (Monogen or TPN) responders with the 8 nonresponders showed no significant difference in age, weight, sex, bilateral pleural effusion, central venous pressure, peak 24-hour volumes, triglycerides, residual lesions, or days from surgery to diagnosis. Fig 1. Median volume of pleural drainage: Monogen (n ⫽ 18) group.
Comment Monogen
died of irreversible pulmonary hypertension following repair of a Taussig Bing anomaly with hypoplastic right lung. Overall, 18 of the original 24 cases (75%) recovered after treatment with Monogen or TPN without need for further surgical intervention. Fourteen of the surviving 17 Monogen patients gained weight from the day of their surgery to hospital discharge. The remaining 3 cases were discharged weighing less than their weight at surgery, although their weight loss was less than 3%. Most of the group went home with Monogen treatment. The interval from the last pleural drain removal to the change to normal formula/diet varied from 1 to 15 weeks. Fifteen patients (68%) stayed on Monogen for 4 ⫾ 1 weeks, and none of the patients reaccumulated a chylothorax after changing back to their normal diet.
Monogen is an MCT-based enteral formula designed for infants and children with lipid and lymphatic disorders. The MCT in both Portagen and Monogen is provided by fractionated coconut oil, but the fatty acid profile of Monogen is unique because it contains 93% MCT with only 7% LCT provided by walnut oil. This LCT level is sufficient to meet essential fatty acid requirements. In contrast, Portagen has 14% of total fat as LCT (4.2 g/L). The osmolality of Monogen is substantially lower than Portagen and most other elemental or fat-free formulas, a difference that improves gastrointestinal tolerance of the feed. Monogen also has a substantially higher energy level than alternative feeds. This point is important because patients with chylothorax have high energy requirements due to increased metabolic demand from the combination of chyle loss and hypermetabolism associated with surgery. We have found that Monogen is an effective form of treatment for postoperative chylothorax in children. Chylothoraces had resolved by 3.5 days in 78% of patients. This rapid response with Monogen compares well with
Monogen Responders Versus Nonresponders A comparison of the 14 Monogen responders with the 4 nonresponders showed no significant difference in age, weight, sex, bilateral pleural effusion, central venous
Table 3. Reported Chylothorax Incidence and Treatment Outcomes in Children Resolved on Enteral Nutrition Cases (n)
Incidence (%)
Survival (%)
Puntis Allen7
15 18
a
1 0.9
93 94
Nguyen3
24
1.9
79
Bond5 Beghetti12
25 51
1.5 2.5
88 90
Current study
25
4.7
92
Study 9
a
Treatment
%
Mean Duration (days)
Low-fat diet ⫹ MCT Low-fat diet (Portagen) Low-fat diet ⫹ TPN Low-fat diet ⫹ MCT (Portagen or Vivonex) or TPN only TPN only Low-fat diet for 1 week then TPN Monogen
73 67
12 11.3
73 39
12 7
66
27
84
15.7
73 80
11.9 24.7
75
Median 5.0
One percent of patent ductus arteriosis procedures.
MCT ⫽ medium-chain triglyceride;
Resolved on Enteral or TPN
TPN ⫽ total parenteral nutrition.
Not stated
Not stated
78
Median 3.5
%
Mean Duration (days)
GENERAL THORACIC
Ann Thorac Surg 2004;77:301–5
304
CORMACK ET AL PEDIATRIC POSTOPERATIVE CHYLOTHORAX
Ann Thorac Surg 2004;77:301–5
GENERAL THORACIC Fig 2. Algorithm of management of chylothorax. (LCT ⫽ long-chain triglyceride.)
other forms of enteral feeding such as Portagen [3, 6, 7], Vivonex [3], and “fat-free formula” [8]. A positive response with reduction of chyle drainage occurred by the end of the third day. Based on our results, in the absence of contraindications to enteral feeding, we believe an initial trial of enteral feeding using Monogen is appropriate. This study also shows that Monogen provides effective nutritional treatment for chylothorax. Body weight in our Monogen group was well maintained with 82% of our patients discharged weighing more than their preoperative weight. Weight loss was minimal in the other 18% (n
⫽ 3), less than 3% of total body weight. More significant weight loss has been seen with other patients with chylothorax when treated with other enteral formulas [7, 9]. Puntis and colleagues [9] reported 8% weight loss in a third of patients. Allen and colleagues [7] reported a higher than 10% weight loss in 22% of their cohort treated with Portagen. We have not found other published reports on the use of Monogen for management of chylothorax. Previous studies have reported use of Portagen [3, 6, 7]. Table 3 summarizes studies using a variety of enteral and parenteral treatments. Several research groups have agreed
that an enteral minimal LCT fat diet is as effective as TPN in controlling lymph leakage in appropriately selected patients [3, 7, 10].
Parenteral Nutrition During this study TPN had been used as initial treatment in 6 patients. The unequal sample size and baseline characteristics of this TPN group and the Monogen group made statistical comparison with the Monogen group potentially misleading. However, the frequent and rapid response with Monogen compares well with this small TPN experience and with others using TPN [5]. Bond and associates [5] found the average duration of drainage was 11.9 days and resolution occurred in 17 of 19 patients, but only after 20 days of total gut rest. The rationale in our ICU for using TPN as the initial treatment was that enteral feeding may be poorly absorbed in the sick neonate or infant, resulting in a poorer nutritional status and the risk of precipitating necrotizing enterocolitis. In a few cases severe fluid restriction may necessitate the use of concentrated TPN to meet nutritional requirements. However, TPN has disadvantages, including gut stasis, catheter-related sepsis, and cost. Our success with Monogen has led us to adjust our initial policy and we now intend to use the formula in ICU patients unless enteral feeding is completely contraindicated. If the pleural drainage is not decreasing by day 5 then TPN should be considered, but investigation for residual cardiac lesions is warranted and early reoperation should be undertaken where appropriate.
Timing of Return to Normal Diet It is generally recommended that patients stay on the MCT formula or minimal LCT fat diet for 3 to 6 weeks, but there is no clear consensus about this strategy or whether to begin timing the treatment from the date of diagnosis, pleural drain removal, finishing TPN, or discharge from hospital [3, 10, 11]. Staying on this restrictive diet longer than necessary is expensive and may mean breast-feeding is not reestablished. This study provides some data about the timing of return to normal diet. None of the patients reaccumulated chyle after they were switched to a normal diet within 6 weeks, nor did the 6 patients who received Monogen for less than 4 weeks. This result confirms 6 weeks on the low LCT diet from the day of the last pleural drain removal is long enough to avoid recurrence. We currently recommend 4 weeks for the duration of Monogen or low LCT diet. An algorithm for the treatment of pediatric postoperative chylothorax based on our findings and review of the literature has now been developed (Fig 2).
Incidence of Chylothorax and Survival The chylothorax incidence rate of 4.7% for the 2-year period in this study is higher than in previous reports. Table 3 shows the reported incidence from five reports published since 1987. Comparison of studies is confounded by varying definitions for the denominator of “total cardiac operations or procedures.” Most authors did not precisely define “total cardiac procedures.” We
CORMACK ET AL PEDIATRIC POSTOPERATIVE CHYLOTHORAX
305
have taken cardiothoracic surgery procedures in patients 10 years of age or younger as the denominator. Some centers may make the diagnosis of chylothorax only when a significant volume or duration is reached. In our study, the mean interval from surgery to diagnosis was 5.9 days in the Monogen group (range 1 to 21 days). This interval is shorter than Le Coultre’s reported mean of 9.8 days and Beghetti’s mean of 7.3 days for those who recovered quickly and 14 days for those with longstanding chylous losses [10, 12]. Earlier recognition and treatment in our group may also help to explain the apparently higher incidence of chylothorax in our unit. The reported incidence appears to increase as more complex surgery is performed [12]. Additionally, a recent trend has been to start feeding patients earlier after surgery, a strategy that may improve nutritional status but that has an unknown influence on the development of chylothorax. With 92% survival, our group compares well with other published survival rates as summarized in Table 3.
Conclusions Monogen is an effective high-energy enteral nutrition treatment for postoperative chylothorax in children. Response is usually rapid within 5 days and body weight is maintained. Treatment can be discontinued after 4 weeks without a recurrence of chylothorax.
References 1. Ramos W, Faintuch J. Nutritional management of thoracic duct fistulas. A comparative study of parenteral versus enteral nutrition. J Parenter Enteral Nutr 1986;10:519 –21. 2. Shaw V, Lawson M. The cardiothoracic system. In: Shaw V, Lawson M, eds. Clinical pediatric dietetics. London: Blackwell Scientific, 2001:185–92. 3. Nguyen DM, Shum-Tim D, Dobell AR, Tchervenkov CI. The management of chylothorax chylopericardium following pediatric cardiac surgery: a 10-year experience. J Card Surg 1995;10:302–8. 4. Marts BC, Naunheim KS, Fiore AC, et al. Conservative versus surgical management of chylothorax. Am J Surg 1992;164:532–4. 5. Bond SJ, Guzzetta PC, Snyder ML, Randolph JG. Management of pediatric postoperative chylothorax. Ann Thorac Surg 1993;56:469 –73. 6. Koloske AM, Martin LW, Schubert WK. Management of chylothorax in children by thoracentesis and medium-chain triglyceride feedings. J Pediatr Surg 1974;9:365–71. 7. Allen EM, van Heeckeren DW, Spector ML, Blumer JL. Management of nutritional and infectious complications of post-operative chylothorax in children. J Pediatr Surg 1991; 26:1169 –74. 8. Buttiker V, Fanconi S, Burger R. Chylothorax in children. Chest 1999;116:682–7. 9. Puntis JW, Roberts KD, Handy D. How should chylothorax be managed? Arch Dis Child 1987;62:593–6. 10. Le Coultre C, Oberhansli I, Mossaz A, Bugmann P, Faidutti B, Belli DC. Postoperative chylothorax in children: difference between vascular and traumatic origin. J Pediatr Surg 1991; 26:519 –23. 11. Bonner G, Warren J. A review of the nutritional management of chyle leakage in adults. J Hum Nutr Diet 1998;11: 105–14. 12. Beghetti M, La Scala G, Belli D, Bugmann P, Kalangos A, Le Coultre C. Etiology and management of pediatric chylothorax. J Pediatr 2000;136:653–8.
GENERAL THORACIC
Ann Thorac Surg 2004;77:301–5