search text   Advanced Search

This Article Abstract Full Text (PDF) P3Rs: Submit a response Alert me when this article is cited Alert me when P3Rs are posted Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Citing Articles Citing Articles via CrossRef Google Scholar Articles by Prais, D. Articles by Harel, L. PubMed PubMed Citation Articles by Prais, D. Articles by Harel, L. Related Collections Infectious Disease & Immunity

Association of Hypoalbuminemia With the Presence and Size of Pleural Effusion in Children With Pneumonia

Department of Pediatrics C, Schneider Children's Medical Center of Israel, Petah Tiqwa and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel

	ABSTRACT

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
OBJECTIVE. Hypoalbuminemia is a common finding in children with massive parapneumonic pleural effusion; however, its incidence and pathogenesis are unclear. The objective of this study was to assess the presence and severity of hypoalbuminemia in children with parapneumonic pleural effusion and to propose a possible pathophysiologic mechanism.

METHODS. The clinical charts of patients who were hospitalized in a tertiary pediatric center with bacterial pneumonia complicated by pleural effusion were reviewed. The volume of pleural fluid was assessed semiquantitatively and categorized as small, moderate, or large. The lowest serum albumin level was recorded, and caloric intake and protein loss were evaluated. Findings were compared with age- and gender-matched children who had bacterial pneumonia without pleural effusion and with children who had acute illnesses other than pneumonia.

RESULTS. Of the 50 patients in the study group, 15 (30%) had small effusions, 16 (32%) had moderate effusions, and 19 (38%) had large effusions. Moderate-to-severe hypoalbuminemia was found in 52% of the study group, 6% of the patients with pneumonia without pleural effusion, and none of the patients with other illnesses. Mean serum albumin level was lower in patients with large pleural effusions than in patients with small effusions (2.66 ± 0.37 vs 3.66 ± 0.47 g/dL). There was no evidence of albumin loss or significant malnutrition. Estimation of the amount of albumin in the drained pleural fluid suggested an albumin shift from blood to pleural fluid.

CONCLUSIONS. Significant hypoalbuminemia is common in children with parapneumonic pleural effusion. Large effusions are associated with low serum albumin levels, which might be explained in part by a shift from blood to pleural fluid.

Key Words: albumin • exudate • pleural effusion • parapneumonia

Parapneumonic fluid is the most common complication of bacterial pneumonia in children¹ and the most common cause of pleural effusion. An estimated 2% of pneumonias are complicated by empyema. Approximately 40% of children who are hospitalized with pneumonia have pleural effusion.²

Pleural effusions can be divided into 2 categories: transudate, resulting from an imbalance between hydrostatic and oncotic pressure, and exudate, resulting from inflammation of the pleura or obstruction of lymphatic flow. This classification, however, was established according to criteria developed and tested in adults,^3,4 and its relevance to children is questionable.⁵

Serum albumin concentration is used as an index of the nutritional status of hospitalized patients. In adults with community-acquired pneumonia, low serum albumin concentration has been found to be a negative prognostic factor for outcome.⁶ Hypoalbuminemia associated with pleural effusion has been attributed to an inflammatory reaction,⁶ protein malnutrition,⁷ or general poor condition.⁸ In the clinical pediatric setting, hypoalbuminemia is recognized as a common feature in acutely infected children with massive pleural effusion. Nevertheless, only a few studies have reported this relationship.^9,10

We hypothesized that children with massive parapneumonic pleural effusion have lower serum albumin levels than other hospitalized children, including children with pneumonia without effusion. The aim of this study was to assess the presence and severity of hypoalbuminemia in hospitalized children with parapneumonic pleural effusion and to propose a possible pathophysiologic mechanism.

	METHODS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The study group consisted of children who had presumed bacterial pneumonia complicated by pleural effusion and attended our tertiary pediatric center from January 2002 to December 2005. Data for the first year of the study were culled retrospectively by review of the clinical records, and data for the subsequent 2 years were collected prospectively. A database was created, and the data were double entered. Diagnostic and therapeutic decisions were based exclusively on clinical criteria. The diagnosis of bacterial pneumonia was established by the presence of fever (>38.5°C), compatible findings on physical examination, elevated levels of acute-phase reactants, and findings of lobar or segmental alveolar infiltrates on chest radiograph. Exclusion criteria for the study were immunodeficiency, malignancy, status posttransplantation, collagen vascular disease, chronic renal disease, surgery during the previous month, and presence of any systemic disease that is known to cause hypoalbuminemia.

The following laboratory results were recorded for the study: complete blood count, erythrocyte sedimentation rate, C-reactive protein level, serum electrolyte levels, renal and liver function tests, and albumin and protein levels. All children underwent upright anterior-posterior and lateral radiographic studies. Radiograph films were interpreted by a pediatric radiologist who was blinded to the patients’ laboratory results. The pleural effusions were graded semiquantitatively according to the radiographic appearance, as follows: small, obliteration of the costophrenic angle and presence of a rim of fluid ascending the lateral chest wall; moderate, opacification of the lower lung field as a result of the presence of fluid; or large, massive opacification of the hemithorax. Chest ultrasonography scans and lateral decubitus films were not performed routinely.

Thoracocentesis was performed in children with a large pleural effusion or respiratory distress. The pleural fluid was investigated for gross appearance, leukocyte count, total protein, albumin, glucose and lactate dehydrogenase levels, pH, and bacterial growth on culture.

For identification of other potential causes of hypoalbuminemia, liver, renal, and coagulation function were measured, in addition to urine protein and sediment and -1-antitrypsin in stool. Caloric intake was determined in general terms by a 24-hour diet recall during the preillness period.

The findings of the study group were compared with 2 control groups. Each patient in the study group was matched for age and gender with 1 child who was hospitalized for lobar or segmental bacterial pneumonia without pleural effusion (control group 1) and 1 child who was hospitalized for an acute illness other than pneumonia (acute gastroenteritis, convulsive episodes, or musculoskeletal skin or viral infection; control group 2). On discharge of a patient in the study group, the first admitted child who met the inclusion criteria for each of the control groups was selected. Patients for whom albumin levels in serum were not measured were excluded from the control groups.

The study protocol was approved by the local institutional review board. Statistical analysis was performed by using BMDP statistical software (University of California Press, Los Angeles, CA). Continuous variables were compared across the groups using analysis of variance with Bonferroni correction for multiple comparisons. Discrete variables were compared by using Pearson's ² test or Fisher's exact test, as appropriate. In the analysis of nonnormally distributed data, square-root transformation was performed before analysis of variance was applied. Multiple stepwise regression was used to test the correlation between predictive variables (magnitude of pleural effusion, albumin) and a dependent variable (length of hospitalization).

	RESULTS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The demographic data of 50 children with parapneumonic pleural effusion and the 2 control groups (n = 50 each) are shown in Table 1. Mean ± SD length of hospitalization was 10.0 ± 4.9 days in the study group, 4.4 ± 1.5 days in the patients with pneumonia without pleural effusion (P < .005), and 3.3 ± 1.3 days in the patients with other illnesses.

Blood cultures were taken from 44 of the 50 children in the study group: 36 were sterile and 8 (18%) grew Streptococcus pneumoniae. S pneumoniae also grew in 6 (15%) cultures from the 42 children tested in control group 1.

Characteristics of the Pleural Effusion
According to the semiquantitative classification, 15 (30%) children in the study group had a small volume of pleural fluid, 16 (32%) had a moderate volume, and 19 (38%) had a large volume. Twenty-five children with pleural effusion underwent thoracocentesis for diagnostic or therapeutic indications (including all of the patients with large pleural effusions and the patients with moderate effusions and respiratory distress). The laboratory characteristics of the pleural fluid samples are shown in Table 2. All patients had exudative fluid: 11 had empyema (frank pus), and 14 had less purulent parapneumonic fluid. There was no association between the size of the effusion and albumin concentration in pleural fluid: small, moderate, and large effusions had similar albumin concentrations, with no statistically significant differences among them.

Serum Albumin Levels
Serum albumin levels were measured for all patients (Fig 1). The patients with pleural effusion had a significantly lower mean value of serum albumin than the other groups (P < .001; Table 3). The lowest values (mean ± SD) during hospital stay were as follows: 3.07 ± 0.58 g/dL in the study group, 3.46 ± 0.34 g/dL in control group 1, and 4.17 ± 0.42 g/dL in control group 2. Fourteen percent of the children with pneumonia, with or without pleural effusion, had levels of serum albumin within the reference range (>3.8 g/dL), compared with 74% of the children who were hospitalized for other illnesses (control group 2). Moreover, 52% of the children in the study group had moderate (2.6–3.0 g/dL) to severe (<2.6 g/dL) hypoalbuminemia as opposed to only 6% of the patients with pneumonia without pleural effusions and none of the patients with other illnesses. This difference was statistically significant (P < .001).

View larger version (14K): [in this window] [in a new window]   FIGURE 1 Serum albumin levels in the different study groups. Each dot represents 1 albumin value.

Serum Albumin Level in Relation to Pleural Fluid Volume
The severity of hypoalbuminemia was found to be related to the volume of pleural effusion. Serum albumin levels were highest in patients with a small volume of pleural effusion (3.66 ± 0.47 g/dL), intermediate in patients with a moderate fluid volume (2.96 ± 0.4 g/dL), and lowest in patients with a large fluid volume (2.66 ± 0.37 g/dL). The difference between the small- and large-volume groups was statistically significant (P < .001; Fig 1).

There was a negative correlation between serum albumin level and days of hospitalization (r = –0.52, P < .001); however, after the volume of pleural effusion was neutralizing, hypoalbuminemia was not an independent predictor of length of hospitalization. Low serum albumin levels were associated with increasing effusion size and, then, with a longer hospital stay. No significant correlation was found between duration of illness and hypoalbuminemia, indicating that the length of illness was not the cause of hypoalbuminemia (r = 0.003).

Other Possible Causes of Hypoalbuminemia
Other potential causes of hypoalbuminemia investigated were albumin loss in urine or the gastrointestinal tract, impaired hepatic synthesis of albumin, and severe malnutrition. Transient proteinuria (>150 mg/dL) was found in 18% of the children with pleural effusion versus 15% in the other groups. None of the children had nephrotic-range proteinuria. Urinalysis findings normalized during hospitalization.

-1-Antitrypsin level in the stool was measured as a marker of protein-losing enteropathy in 32 study group patients. Only 1 patient (with mild hypoalbuminemia) had a slightly elevated level (2.1 mg/g fecal dry weight; reference: <2 mg/g).

Synthetic liver function was assessed by measurement of prothrombin time and calculation of the international normalized ratio. All patients tested in the reference. Mild and transient elevation of liver transaminase levels was observed in 8.3% of the study group and 5% of control group 1.

Assessment of the study group's preillness energy intake was performed by a clinical dietitian. No evidence of chronic malnutrition was found in any of the children. All patients had a body weight above the third percentile for age (Table 1).

Potential Underlying Mechanism: Representative Cases
On the basis of calculations of the albumin deficit and the magnitude of albumin drained by thoracocentesis, we hypothesized that the large effusions that were found to be associated with low serum albumin level may be at least partly explained by a shift in albumin from blood and pleural fluid. Table 4 summarizes 5 cases in which these calculations were performed. All of these patients underwent a single thoracocentesis drainage without insertion of a chest tube. A representative case is described in detail next.

	DISCUSSION

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
This study compared the serum albumin level of hospitalized children who had pneumonia and pleural effusion with that of children who had pneumonia without effusion and with children who had acute illnesses other than pneumonia. We found that hypoalbuminemia was a common feature in patients with parapneumonic pleural effusion, with 52% of our study group showing moderately to severely low albumin levels. Moreover, the greater the magnitude of the pleural fluid, the lower the albumin level. Hypoalbuminemia was more frequent and more severe in the study patients than in the other 2 groups. Because the diagnostic and therapeutic decisions during hospitalization were based only on clinical criteria, this study can be considered as a "true to life" analysis. We suggest that in this clinical setting, the presence of hypoalbuminemia might serve as a subtle clue of an evolving pleural effusion (especially when chest ultrasonography is not immediately available.)

The association between low serum albumin levels and pleural effusion is commonly recognized in the pediatric clinical setting, but it has seldom been investigated or reported.^9,10 In adults, it has usually been reported in patients with chronic diseases (eg, congestive heart failure, malignancy), for whom the hypoalbuminemia is usually the cause of the pleural effusion.¹¹ Several researchers proposed that the lower serum albumin levels in patients with pleural effusion may also indicate a poor general condition⁸ or undernutrition.⁷ Hypoalbuminemia was also reported to correlate with poor outcome in adult patients with recurrent pneumonia¹² and with a higher mortality risk.^13,14 We did not measure serum immunoglobulin levels; however, concomitant hypogammaglobulinemia with hypoalbuminemia could provide an additional explanation for the poor outcome in this population. Because of the inverse correlation found between albumin and acute-phase protein levels in adults who were hospitalized with pneumonia, some authors suggested that the inflammatory process itself was the main reason for the hypoalbuminemia.⁶

In this study, in addition to the patients with parapneumonic pleural effusion, some degree of hypoalbuminemia was found in the patients who had pneumonia without pleural effusion and in a small proportion of patients who were hospitalized for acute illnesses other than pneumonia. Among the children with pleural effusion, serum albumin level generally decreased with increasing effusion size, but the serum albumin levels were similar in control group 1 and in the small- and moderate-effusion groups (Fig 1). The possible causes of these low levels of albumin may have been inflammation, low energy intake, decreased protein synthesis, and catabolic state, although neither liver dysfunction nor protein losses were found. Thus, serum albumin levels can probably be used as a marker of disease severity in patients with pneumonia, even without pleural effusion; however, in children with large parapneumonic pleural effusion, we propose an additional explanation for the cause of hypoalbuminemia, namely, an albumin shift from the intravascular space to the pleural fluid. A similar hypothesis was suggested by others in a single case report.⁹

In the cases described in Table 4, we calculated that the quantity of albumin present in the pleural space could feasibly account for the loss of serum albumin. Obviously, our calculations were theoretical and inexact, especially owing to the gross estimation of the pleural fluid volume based on a simple 1-time thoracocentesis. Nevertheless, these findings indicate a similar magnitude of intravascular albumin deficit and albumin concentration in the pleural space. Although we could not determine the precise burden of the albumin shift, we can assume that it is an important factor in the pathophysiology of significant hypoalbuminemia in children with parapneumonic pleural effusion. The positive correlation found between the amount of pleural effusion and the severity of hypoalbuminemia supports this hypothesis. Furthermore, the serum-to-effusion albumin gradient has been shown to be sensitive in distinguishing transudative from exudative effusions (a gradient of >1.2 g/dL is indicative of transudate).¹⁵ Our hypothesis is therefore consistent with previous reports that the low serum albumin levels were attributable to shifting from blood to pleural fluid.^9,15

In our study group, other probable causes of hypoalbuminemia seem unlikely, namely, urinary or fecal loss (normal urinalysis results and -1-antitrypsin in stool), malnutrition (weight percentiles and nutritional assessment), and significantly impaired synthetic liver function (normal coagulation studies). Klar et al¹⁰ reported a similar rate of hypoalbuminemia in children with pneumonia and pleural effusion, but they found an elevated level of fecal -1-antitrypsin in 59% of their patients and consequently suggested protein-losing enteropathy as a potential cause of the hypoalbuminemia. We could not confirm this observation.

Another hypothesis that should be considered is that the sickest patients are those with the largest effusions and the lowest serum albumin, although, in this study, hypoalbuminemia was not an independent predictive factor of length of hospitalization. Theoretically, the presence of proteins in pleural effusion, including immunoglobulins, could play a role in the pathologic process. This and other possible pathophysiologic mechanisms need to be proposed and analyzed in future studies.

The bacteriologic identification of the culprit pathogen was clearly enhanced by pleural tap. Similar to previous reports,⁵ bacteremia was identified in 18% of the study patients, and pleural culture was positive in 36%. In only 1 patient were both blood and pleural cultures positive. The reported rates of identification of infectious organisms from pleural fluid vary markedly from 8% to 76%.⁵ Today, the early introduction of antibiotic treatment tends to lower the rate of culture-positive samples.

This study had some limitations, including the relatively small sample size and imprecise estimation of pleural fluid volume by chest radiograph studies, without a quantitative estimation of the pleural fluid volume by chest ultrasound. Moreover, longitudinal measurements of serum albumin levels were not performed. We performed a "true to life" study, without executing any diagnostic procedures other than those clinically indicated. Prospective studies of the kinetics of serum albumin in relation to onset of disease and appearance of pleural effusion could yield relevant results, including the possibility that hypoalbuminemia should guide the type of therapeutic or diagnostic intervention. Likewise, additional studies are needed to determine whether high protein intake ameliorates the clinical status of patients with parapneumonic pleural effusion or merely increases the rate of albumin shift into the pleural space. Nevertheless, our findings are important primarily because they contribute to the knowledge on the pathophysiology of hypoalbuminemia associated with parapneumonic pleural effusion and also because of their clinical significance, namely, that the presence of hypoalbuminemia in children with lobar pneumonia might be the first subtle clue of an incipient pleural effusion.

	CONCLUSIONS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Significant hypoalbuminemia is a very common finding in children with parapneumonic pleural effusion. Lower albumin levels correlate with greater magnitudes of pleural fluid volume. An albumin shift from the serum to the pleura is probably an important factor in the pathophysiology of hypoalbuminemia in this setting.

	ACKNOWLEDGMENTS

 
We thank Gloria Ginzach and Phyllis Curchack Kornspan for assistance.

	FOOTNOTES

 
Accepted Jul 27, 2007.

Address correspondence to Dario Prais, MD, Department of Pediatrics C, Schneider Children's Medical Center of Israel, Petah Tiqwa 49202, Israel. E-mail: dariop{at}clalit.org.il

The authors have indicated they have no financial relationships relevant to this article to disclose.

This work was presented in part at the European Respiratory Society Annual Congress; September 2–6, 2006; Munich, Germany.

	REFERENCES

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 

1. Alkrinawi S, Chernick V. Pleural infection in children. Semin Respir Infect.1996; 11 (3):148 –154[Medline]
2. Mocelin HT, Fischer GB. Epidemiology, presentation and treatment of pleural effusion. Paediatr Respir Rev.2002; 3 (4):292 –297[CrossRef][Medline]
3. Light RW. Pleural effusion. N Engl J Med.2002; 346 (25):1971 –1977[Free Full Text]
4. Bye MR. Identifying the cause of pleural effusion in children. J Respir Dis.1982; 3 (7):13 –22
5. Balfour-Lynn IM, Abrahamson E, Cohen G, et al. BTS guidelines for the management of pleural infection in children. Thorax.2005; 60 (suppl 1):i1 –i21[Free Full Text]
6. Hedlund JU, Hansson LO, Ortqvist AB. Hypoalbuminemia in hospitalized patients with community-acquired pneumonia. Arch Intern Med.1995; 155 (13):1438 –1442[Abstract]
7. Sullivan DH, Sun S, Walls RC. Protein-energy undernutrition among elderly hospitalized patients. JAMA.1999; 281 (21):2013 –2019[Abstract/Free Full Text]
8. Mattison LE, Coppage L, Alderman DF, Herlong JO, Sahn SA. Pleural effusions in the medical ICU: prevalence, causes and clinical implications. Chest.1997; 111 (4):1018 –1023[CrossRef][ISI][Medline]
9. Porat S, Bhatia N, Barnett DW. A child with severe pneumonia, pleural effusion and acute hypoalbuminemia. Clin Pediatr (Phila).2002; 41 (3):199 –200[Free Full Text]
10. Klar A, Shoseyov D, Berkun Y, et al. Intestinal protein loss and hypoalbuminemia in children with pneumonia. J Pediatr Gastroenterol Nutr.2003; 37 (2):120 –123[CrossRef][Medline]
11. Eid AA, Keddissi JI, Kinasewitz GT. Hypoalbuminemia as a cause of pleural effusions. Chest.1999; 115 (4):1066 –1069[CrossRef][ISI][Medline]
12. Hedlund JU, Ortqvist AB, Kalin ME, Granath F. Factors of importance for the long term prognosis after hospital treated pneumonia. Thorax.1993; 48 (8):785 –789[Abstract]
13. Goldwasser P, Feldman J. Association of serum albumin and mortality risk. J Clin Epidemiol.1997; 50 (6):693 –703[CrossRef][ISI][Medline]
14. Afessa B, Greaves WL, Fredercik WR. Pneumococcal bacteremia in adults: a 14-year experience in an inner-city university hospital. Clin Infect Dis.1995; 21 (2):345 –351[ISI][Medline]
15. Segura RS. Useful clinical biological markers in diagnosis of pleural effusions in children. Paediatr Respir Rev.2004; 5 (suppl A):S205 –S212[CrossRef][Medline]

This Article Abstract Full Text (PDF) P3Rs: Submit a response Alert me when this article is cited Alert me when P3Rs are posted Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Citing Articles Citing Articles via CrossRef Google Scholar Articles by Prais, D. Articles by Harel, L. PubMed PubMed Citation Articles by Prais, D. Articles by Harel, L. Related Collections Infectious Disease & Immunity

	Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2008 American Academy of Pediatrics. All rights reserved.