Klar, Aharon*,§; Shoseyov, David‡; Berkun, Yaakov§; Brand, Abraham†; Braun, Jacques§; Shazberg, Gila§; Jonathan, Moise§; Gross-Kieselstein, Eva§; Revel-Vilk, Shoshana§; Hurvitz, Haggit§
*Gastroenterology, †Cardiology, and ‡Pulmonology Units of the §Department of Pediatrics, Bikur Cholim General Hospital, affiliated with the Hebrew University-Hadassah Medical School, Jerusalem, Israel
Received April 19, 2002; accepted February 28, 2003.
Address correspondence and reprint requests to Dr. Aharon Klar, Gastroenterology Unit, Department of Pediatrics, Bikur Cholim General Hospital, POB 492, Jerusalem, 91004, Israel (e-mail: firstname.lastname@example.org).
Background: Intestinal protein loss has been reported mainly in diseases affecting the gastrointestinal tract. Intestinal protein loss during pneumonia with effusion has not been reported to date. The authors attempted to assess the associations between pneumonia with effusion and intestinal protein loss and hypoalbuminemia in children.
Methods: This was a prospective consecutive case series study of in children hospitalized with pneumonia and effusion during a period of 4½ years. Serum albumin, C-reactive protein (CRP), and fecal α-1 antitrypsin (α-1-AT) were measured in the first 72 hours of hospitalization. Two control groups were studied: one consisted of 50 febrile children hospitalized because of viral or mild bacterial infections, and the other consisted of 20 afebrile children hospitalized because of convulsive disorders.
Results: Sixty-seven children ages 4 months to 14 years hospitalized with pneumonia and effusion were enrolled in the study. Fifty-nine percent (40 children) were found to have elevated fecal α-1-AT excretion (range, 2–10 mg/g) compared with none in the two control groups (P < 0.000).
Fifty-two percent (35 children) of the children with pneumonia and effusion had mild to moderate hypoalbuminemia (range, 22–34 g/L). Only one child (2%) in the febrile control group had a low albumin of 34 g/L; none were found in the afebrile control group. The level of fecal α-1-AT was inversely correlated with serum albumin level.
Conclusions: Pneumonia with effusion in children is often associated with an intestinal protein loss that can be monitored by measuring gastrointestinal loss of protein, namely fecal α-1-AT. In most cases the development of hypoalbuminemia correlates with the development of intestinal protein loss.
Gastrointestinal manifestations of pneumonia may include abdominal pain, paralytic ileus, vomiting, and diarrhea. An intestinal protein loss during pneumonia with effusion has not been previously described. In adults, hypoalbuminemia during pneumonia has been described and attributed to extracellular protein redistribution and peripheral protein catabolism (1).
The association of pneumonia with protein-losing enteropathy has never been reported. In this prospective study, we investigate children with pneumonia and effusion for protein-losing enteropathy and hypoalbuminemia.
PATIENTS AND METHODS
A prospective study was conducted during a period of 4.5 years from July 1996 to December 2000 in our university-affiliated pediatric ward. The study group included all children hospitalized with severe pneumonia and effusion from whom blood and stool samples were taken during the first 72 hours of hospitalization.
All patients underwent chest x-ray, ultrasound, and electrocardiography; echocardiography was performed when indicated. Ultrasonography of the gastric folds (ATL Ultramark) with HDI linear transducer of 5 MHz, Bothell, WA) was performed in 15 children with the most severe hypoalbuminemia. Stools were examined for Giardia lamblia and other parasites, by direct microscopy. Two control groups were studied. The first group consisted of 20 afebrile children (ages 1–13 years) hospitalized because of epilepsy or fainting (afebrile group). The second group consisted of 50 febrile children (ages 4 months to 12 years) with viral disease or mild bacterial disease: pharyngitis, otitis media, or cellulitis (fever group). In all patients, serum levels of albumin and C-reactive protein (CRP) were examined. Stool was examined for fecal α-1 antitrypsin (α-1-AT). Albumin and CRP levels were analyzed by routine biochemical procedures and fecal α-1-AT by the random fecal α-1-AT test (2), the nephelometric technique (Beckman USA); normal levels are less than 2 mg/g fecal dry weight. Informed consent for fecal α-1-AT testing was obtained from all parents.
None of the children in the study group had known gastrointestinal liver or renal disease. Cardiac abnormalities were found in three children: one atrial septal defect, one atrial septal defect and patent ductus arteriosus, and one ventricular septal defect. None of these three children had evidence of congestive heart failure.
Statistical analysis was performed using GB-stat software on a PC computer; ANOVA was performed to compare the levels of serum albumin and fecal α-1-AT of the three groups. Regression analysis was performed to assess the correlation between serum albumin and fecal α-1-AT.
Sixty-seven children ages 4 months to 14 years were studied. All gave stool and blood samples during the first 72 hours of hospitalization. The study group (pneumonia) mean age of 3.6 years ± 2.9 years and the febrile group (viral) ages of 4 months to 12 years (mean 3.4 years ± 3.2 years) were matched. The children in the afebrile group 1 to 13 years were older than the study group mean of 6.7 years ± 4.2 years; however, fecal α-1-AT excretion is not age dependent after the neonatal period (3).
Forty (59%) children with pneumonia and effusion were found to have elevated fecal α-1-AT excretion (range, 2–10 mg/g fecal dry weight). The fecal α-1-AT level for the whole group was 0.5 to 10 mg/g (mean, 3.53 ± 2.45 mg/g). No child in the two control groups had elevated fecal α-1-AT excretion. The range of fecal α-1-AT excretion in the febrile group was 0.5 to 1.6 mg/g (mean, 0.61 ± 0.3 mg/g and 0.5 ± 0 mg/g in the afebrile group). P < 0.000 for both groups (0.5 mg/g was the lowest measurable level in the laboratory).
Hypoalbuminemia of less than 34 g/L (range, 22–34 g/L) was found in 35 (52%) children in the study (pneumonia) group (range, 22–46 g/L; mean, 33.6 ± 5.6 g/L for the whole group), versus 2% in the febrile control group (albumin range, 34–49 g/L; mean, 41 ± 3.5 g/L; P < 0.000). No case of hypoalbuminemia was found in the afebrile group (albumin range, 38–50 g/L; mean, 41.7 ± 3.6 g/L; P < 0.000). Regression analysis of serum albumin and fecal α-1-AT demonstrated inverse correlation between them.
CRP levels were between 89 and 410 mg/L (mean, 252 ± 97 mg/L; N < 5 mg/L) in children with pneumonia and effusion; 2 and 46 mg/L (mean, 18 ± 14 mg/L) in the febrile control group (P <0.000); and 0.5 and 3 mg/L (mean, 1.4 ± 1.1 mg/L) in the afebrile group. Complete data on 11 children with pneumonia who gave sequential blood and stool samples are reported in Table 1.
The development of a transient intestinal protein loss in children with pneumonia and effusion has not previously been described. This protein loss was monitored by measuring fecal α-1-AT excretion.
Elevated excretion of fecal α-1-AT was accompanied in many cases by hypoalbuminemia.
The presence of high fecal α-1-AT levels could not be attributed to swallowed proteinaceous secretions related to the pneumonia because this protein is acid sensitive and is destroyed during gastric passage (3).
Transient protein-losing enteropathy has been described in children with various infectious and noninfectious gastrointestinal conditions (4–8). The most common ones, Menetrier disease, and parasitic gastrointestinal infection, were ruled out in our patients by their clinical presentation, ultrasound of the gastric folds, the patients' rapid spontaneous recovery, and their stool examinations, which were negative for parasites and leukocytes.
Antibiotic-associated colitis (9) as a cause of hypoalbuminemia was ruled out on the basis of clinical and laboratory evaluation.
There are rare reports of excessive protein loss into the gastrointestinal tract in patients with congestive heart failure, mostly constrictive pericarditis, or after surgical correction of congenital cardiac anomalies (10–12). Although cardiac abnormalities were found in three children, there was no evidence of congestive heart failure in any of these patients.
Congenital anomalies of the lymphatic drainage from the gastrointestinal tract causing protein-losing enteropathy have been reported in a child with the syndrome of yellow nails, lymphedema, and pleural effusion (13). However, in children without a preexisting lymphatic disorder, the development of protein-losing enteropathy in the course of pneumonia with effusion has, to the best of our knowledge, never been reported.
In adults, the mechanism responsible for the development of hypoalbuminemia during injury and infection has been studied by several investigators. Dahn et al. suggested that an extravascular protein redistribution, sometimes together with increased peripheral protein catabolism, is the major factor responsible for hypoalbuminemia (14). Hedlund and associates suggested that the low serum albumin level results from the inflammatory process per se (15). Protein undernutrition in elderly hospitalized patients was recently reported as contributing to hypoalbuminemia (16). Merritt et al. studying the significance of hypoalbuminemia in pediatric oncology, concluded that low albumin may reflect a response to infection, rather than nutritional status and depletion of body mass (17).
A recent review paper dealing with acute phase proteins lists albumin as a protein whose plasma concentration decreases during diseases involving inflammation. However, no explanation is provided for the development of hypoalbuminemia during inflammation, other than reduced production (18).
Hennig et al. studied the pathogenesis of hypoalbuminemia and found that administration of tumor necrosis factor (TNF-α) to healthy, well-nourished rabbits led to hypoalbuminemia. They also showed that TNF-α increased transendothelial passage of albumin in vitro (19).
On the basis of these observations, we suggest that in patients with severe infections the infectious process induces the release of cytokines such as TNF-α, interleukin-1, and interleukin-6 that increase intestinal permeability, leading to loss of protein (19–23). This protein loss may be aggravated by pulmonary hypertension and increased right atrial pressure in patients with pneumonia (24). These hypotheses require substantiation by future studies.
Because many proteins, including immunoglobulins, may leak into the gastrointestinal tract in the protein-losing process, elucidation of the mechanisms leading to protein-losing enteropathy may indicate therapeutic measures for reducing mortality in adults and morbidity in children. Studies on the role of protein-losing enteropathy in other systemic diseases are in progress.
Our report expands the differential diagnosis of protein-losing enteropathy to include disorders not directly affecting the gastrointestinal system. We suggest adding pneumonia with effusion to the growing list of diseases that can cause transient protein-losing enteropathy in children.
The authors thank Mrs. Ilana Ivas, head nurse, and the staff of the department of Pediatrics for their assistance, and Mrs. Nava Alon for her excellent secretarial assistance.
1. Afessa B, Greaves WL, Frederick WR. Pneumococcal bacteremia in adults: a 14-year experience in an inner-city university hospital. Clin Infect Dis 1995; 21:345–51.
2. Magazzu G, Jacono G, Di Pasquale G, et al. Reliability and usefulness of random fecal alpha 1-antitrypsin concentration: further simplification of the method. J Pediatr Gastroenterol Nutr 1985; 4:402–7.
3. Wallace A, Gleason JR. Protein losing enteropathy. In: Wyllie R, Hyams JS, eds. Pediatric Gastrointestinal Disease. Pathophysiology, Diagnosis, Management. Philadelphia: WB Saunders; 1993:537–8.
4. Thomas DW, Sinatra FR, Merritt RJ. Random alpha-1-antitrypsin concentration in children with gastrointestinal disease. Gastroenterology 1981; 80:776–82.
5. Cieslak TJ, Mullett CT, Puntel RA, et al. Menetrier's disease associated with cytomegalovirus infection in children: report of two cases and review of the literature. Pediatr Infect Dis J 1993; 12:340–3.
6. Korman SH, Bar Oz B, Mandelberg A, et al. Giardiasis with protein-losing enteropathy: diagnosis by fecal alpha 1-antitrypsin determination. J Pediatr Gastroenterol Nutr 1990; 10:249–52.
7. Reif S, Jain A, Santiago J, et al. Protein losing enteropathy as a manifestation of Henoch-Schönlein purpura. Acta Paediatr Scand 1991; 80( 4):482–5.
8. Wood ML, Foulds IS, French MA. Protein losing enteropathy due to systemic lupus erythematosus. Gut 1984; 25:1013–5.
9. Pickering LK, Cleary TG. Gastrointestinal infections (section six): approach to patients with gastrointestinal tract infections and food poisoning. In: Reigin RD, Cherry JD, eds. Textbook of Pediatric Infectious Diseases.
3rd ed. Philadelphia: WB Saunders; 1992:568.
10. Petersen VP, Mastrup J. Protein-losing enteropathy in constrictive pericarditis. Acta Med Scand 1963; 173:401–10.
11. Jacobs ML, Rychik J, Byrum CJ, et al. Protein-losing enteropathy after Fontan operation: resolution after baffle fenestration. Ann Thorac Surg 1996; 61:206–8.
12. Berkowitz I, Segal I. Protein-losing enteropathy in congestive cardiac failure: an entity of minor clinical significance. Am J Gastroenterol 1990; 85:154–6.
13. Battaglia A, di Ricco G, Mariani G, et al. Pleural effusion and recurrent broncho-pneumonia with lymphedema, yellow nails and protein-losing enteropathy. Eur J Respir Dis 1985; 66:65–9.
14. Dahn MS, Jacobs LA, Smith S, et al. The significance of hypoalbuminemia following injury and infection. Am Surg 1985; 51:340–3.
15. Hedlund JU, Hansson LO, Ortqvist AB. Hypoalbuminemia in hospitalized patients with community-acquired pneumonia. Arch Intern Med 1995; 155:1438–42.
16. Sullivan DH, Sun S, Walls RC. Protein-energy undernutrition among elderly hospitalized patients: a prospective study. JAMA 1999; 281:2013–9.
17. Merritt RJ, Kalsch M, Roux LD, et al. Significance of hypoalbuminemia in pediatric oncology patients–malnutrition or infection. J Parenter Enteral Nutr 1985; 9:303–6.
18. Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med 1999; 340:448–53.
19. Hennig B, Honchel R, Goldblum SE, et al. Tumor necrosis factor-mediated hypoalbuminemia in rabbits. J Nutr 1988; 118:1586–90.
20. McClain CJ, Hennig B, Ott LG, et al. Mechanisms and implications of hypoalbuminemia in head-injured patients. J Neurosurg 1988; 69:386–92.
21. Gilmont RR, Dardano A, Engle JS, et al. TNF-alpha potentiates oxidant and reperfusion-induced endothelial cell injury. J Surg Res 1996; 61:175–82.
22. Boirivant M, Pallone F, Ciaco A, et al. Usefulness of fecal alpha 1-antitrypsin clearance and fecal concentration as early indicator of postoperative asymptomatic recurrence in Crohn's disease. Dig Dis Sci 1991; 36( 3):347–52.
23. Picco P, Gattorno M, Marchese N, et al. Increased gut permeability in juvenile chronic arthritides. A multivariate analysis of the diagnostic parameters. Clin Exp Rheumatol 2000; 18( 6):773–8.
24. Jun-Bao D, Shu-Zheng L, Bao-Lin W, et al. Doppler echocardiographic evaluation of pulmonary artery pressure in pneumonia of infants and children. Pediatr Pulmonol 1991; 10:296–8.
© 2003 Lippincott Williams & Wilkins, Inc.