Characterization of hypoalbuminemia following temporary abdominal closure

Abstract

BACKGROUND

The purpose of this study was to characterize associations among serum proteins, negative-pressure wound therapy (NPWT) fluid loss, and primary fascial closure (PFC) following emergent laparotomy and temporary abdominal closure (TAC). We hypothesized that high levels of C-reactive protein (CRP) and NPWT output would be associated with hypoalbuminemia and failure to achieve PFC.

METHODS

We performed a retrospective analysis of 233 patients managed with NPWT TAC. Serum proteins and resuscitation indices were assessed on admission, initial laparotomy, and then at 48 hours, 96 hours, 7 days, and discharge. Correlations were assessed by Pearson coefficient. Multivariable regression was performed to identify predictors of PFC with cutoff values for continuous variables determined by Youden index.

RESULTS

Patients who failed to achieve PFC (n = 55) had significantly higher CRP at admission (249 vs. 148 mg/L, p = 0.003), initial laparotomy (237 vs. 154, p = 0.002), and discharge (124 vs. 72, p = 0.003), as well as significantly lower serum albumin at 7 days (2.3 vs. 2.5 g/dL, p = 0.028) and discharge (2.5 vs. 2.8, p = 0.004). Prealbumin (in milligrams per deciliter) was similar between groups at each time point. There was an inverse correlation between nadir serum albumin and total milliliters of NPWT output (r = −0.33, p < 0.001). Exogenous albumin administration (in grams per day) correlated with higher serum albumin levels at each time point: 48 hours: r = 0.26 (p = 0.002), 96 hours: r = 0.29 (p = 0.002), 7 days: r = 0.40 (p < 0.001). Albumin of less than 2.6 g/dL was an independent predictor of failure to achieve PFC (odds ratio, 2.59; 95% confidence interval, 1.02–6.61) in a multivariate model including abdominal sepsis, body mass index of greater than 40 kg/m², and CRP of greater than 250 mg/L.

CONCLUSIONS

Early and persistent systemic inflammation and high NPWT output were associated with hypoalbuminemia, which was an independent predictor of failure to achieve PFC. The utility of exogenous albumin following TAC requires further study.

Large volumes of albumin are lost when protein-rich peritoneal fluid is removed by negative-pressure wound therapy (NPWT) temporary abdominal closure (TAC) dressings.¹ Patients with TAC are often affected by hypovolemic or distributive shock and capillary leak, which allows albumin to extrude to the interstitial space, creating a pathophysiologic oncotic gradient.^2,3 Correction of hypoalbuminemia is hindered by persistent systemic inflammation, which suppresses hepatic albumin synthesis.^4–6 Hypoalbuminemia may then propagate visceral edema and failure to achieve primary fascial closure (PFC). Delays in PFC perpetuate NPWT fluid losses and multiple operations to cover and protect the viscera, and the cycle continues. Therefore, hypoalbuminemia following TAC may be a clinically significant phenomenon, but has not been fully characterized.

Better understanding of the relationship between hypoalbuminemia and PFC may by incorporated with previously identified risk factors for adverse events and failure to achieve fascial closure, including bowel resection, large-volume fluid resuscitation, multiple reoperations, prolonged open-abdomen therapy, and delays in providing enteral nutrition.^7–10 The purpose of this study was to assess associations among systemic inflammation, serum albumin, NPWT fluid loss, and PFC. We hypothesized that high levels of C-reactive protein (CRP) and NPWT output would correlate with hypoalbuminemia and that hypoalbuminemia would be associated with failure to achieve PFC.

METHODS

Design and Participants

This study was a retrospective analysis of 233 consecutive TAC patients treated at our institution between June 2012 and June 2016. Institutional review board approval was obtained. Patients were identified by searching our database for adult (aged ≥18 years) patients who had Current Procedural Terminology codes for relaparotomy (49002) or exploratory laparotomy (49000) with a planned or unplanned reoperation Current Procedural Terminology modifier (48, 78). Patients meeting these criteria were included if they underwent TAC with an NPWT dressing. Patients who survived for less than 96 hours following initial TAC were excluded to ensure that all subjects had the opportunity to achieve early PFC and to allow for the analysis of trends over time for key parameters. Dialysis patients were excluded to avoid the confounding effects of glomerular filtration rate on albumin and prealbumin serum levels and decreased albumin synthesis among dialysis patients.^11,12 Patients with cirrhosis were excluded to avoid the confounding effects of cirrhosis on NPWT output and serum protein levels.^13,14 Patients with necrotizing pancreatitis were excluded because of significant differences in the preoperative and postoperative courses associated with this disease process. Derivation of the study population is illustrated in Supplemental Digital Content 1 (see Figure, Supplemental Digital Content 1, http://links.lww.com/TA/B33).

Although trauma and nontrauma patients managed by TAC are distinct populations with clinically important differences in resuscitation requirements and operative course,¹⁵ the weight of evidence suggests that the systemic inflammatory process and capillary leak in these populations are similar.^16–19 In addition, a multivariable logistic regression model should ideally contain at least 10 outcome events for each variable in the model.²⁰ In our model, there were 55 total cases of failure to achieve fascial closure and four variables in the model. However, there were only nine cases of failure to achieve fascial closure for trauma patients, such that a subgroup analysis of trauma patients may be underpowered and overfit.

Procedures and Parameters

During the study period, postoperative resuscitation was protocolized as previously described.^21–23 Operative techniques were at the discretion of the attending surgeon. All patients were managed with NPWT TAC and intention for planned relaparotomy and sequential fascial closure attempts at 24-hour to 48-hour intervals, favoring relaparotomy within 24 hours if possible.²⁴ Negative-pressure wound therapy included use of individualized vacuum pack (i.e., Barker technique²⁵) and commercial (ABThera System, KCI, San Antonio, TX) vacuum-assisted closure dressings. If visceral edema or critical loss of abdominal domain precluded PFC during the index hospital admission, a planned ventral hernia was created with biologic mesh or polyglactin mesh with interval split-thickness skin grafting. All clinical parameters were derived from our institutional database or recorded by review of the electronic medical record.

Trends in serum proteins and resuscitation indices were assessed on admission, at initial exploratory laparotomy and TAC, and following initial TAC at 48 hours, 96 hours, 7 days, and discharge. C-reactive protein (in in milligrams per liter), albumin (in grams per deciliter), and prealbumin (in milligrams per deciliter) values were obtained from laboratory assessments performed as standard care, often for nutritional assessment. C-reactive protein, albumin, and prealbumin values were available at the time of TAC for 77, 191, and 54 subjects, respectively. The numbers of available values for CRP, albumin, and prealbumin at all time points are illustrated in Figure 1. The numbers of missing values for all other variables are listed in Supplemental Digital Content 2 (see Table, Supplemental Digital Content 2, http://links.lww.com/TA/B34). All patients in the study population were included in the analysis, and missing values were not imputed or replaced. It is possible that missing values for lactate, albumin, and prealbumin may have been influenced by clinical context; that is, lactate levels may not have been assessed for patients in whom there was low suspicion for hypoxia and ischemia, and albumin and prealbumin levels may not have been assessed for patients in whom the magnitude of nutritional deficiency or inflammatory response was not in question. Because all laboratory measurements were performed as standard care assessments, it is difficult to determine retrospectively the percentage of each value missing at random versus missing not at random. It is also possible that values for intravenous fluid administration, exogenous albumin administration, NPWT output, and transfusions may have been missing at random if they were not recorded properly in the electronic medical record, as the medical record will autopopulate a value of zero when there is no recorded input.

Figure 1.

Trends in CRP (A), serum albumin (B), and prealbumin (C) over time for patients undergoing exploratory laparotomy and TAC. Square points and dotted lines represent patients who achieved PFC; circle points and solid lines represent patients who did not achieve PFC during admission. DC indicates discharge. *p < 0.03 between groups.

Statistical Analysis

Statistical analysis was performed in SPSS (version 23; IBM, Armonk, NY). Continuous variables describing the study population were reported as median (interquartile range). Serum levels of CRP (in milligrams per liter), albumin (in grams per deciliter), and prealbumin (in milligrams per deciliter) followed normal distributions and were graphically represented (GraphPad Prism, version 6.05; GraphPad Software, La Jolla, CA) by mean values with 95% confidence intervals (95% CIs) and compared by one-way analysis of variance. Discrete variables were compared by Fisher exact test and reported as n (%). The relationship between serum albumin and NPWT output and relationships between exogenous albumin and serum albumin were assessed by Pearson correlation coefficient. Independent predictors of PFC were identified on multivariable logistic regression. Model covariates were identified by assessing bivariate correlations between salient parameters and PFC. Variables were included if they had significant correlation to fascial closure (r > 0.2, p < 0.05) and lacked significant colinearity to the other covariates. Optimal cutoff values for continuous variables were assessed by Youden index,²⁶ calculating the point at which the combination of sensitivity and specificity was greatest. Model strength was assessed by calculating the area under the receiver operating characteristic curve.

RESULTS

Patient Characteristics and Outcomes

Two hundred thirty-three patients were included (Table 1). The study population was predominantly middle aged (median age, 56 years) with a low burden of chronic disease (Charlson Comorbidity Index 1.0) and incapacitating systemic illness at the time of initial laparotomy (American Society of Anesthesiologists class 4.0). Sixty-one percent had a primary diagnosis of abdominal sepsis, which was most commonly due to bowel ischemia. Patients who presented with intra-abdominal sepsis had a significantly greater number of abdominal operations (3.0 [2.0–4.0]) compared with all other patients (2.0 [2.0–3.0], p = 0.021). Trauma patients composed 34%of the study population, and 20%of all trauma patients had penetrating injuries. Primary fascial closure was achieved in 89% of all trauma patients and 70% of all nontrauma patients (p = 0.001). Patients who achieved fascial closure had lower crystalloid resuscitation volumes following laparotomy (Table 2), although the differences were not statistically significant. Negative-pressure wound therapy fluid losses were significantly higher among patients who did not achieve PFC. Overall inpatient mortality was 12%.

TABLE 1.

Characteristics of the Study Population

Patient Characteristics	All Patients (n = 233)	PFC (n = 178)	No PFC (n = 55)	p
Age, y	56 (43–69)	56 (40–70)	57 (48–65)	0.630
Male	123 (53%)	96 (54%)	27 (49%)	0.541
Charlson Comorbidity Index	1.0 (0.0–3.0)	1.0 (0.0–3.0)	2.0 (0.0–3.0)	0.060
ASA physical status classification	4.0 (3.0–4.0)	4.0 (3.0–4.0)	4.0 (3.0–4.0)	0.602
Body mass index, kg/m2	28 (23–33)	27 (23–32)	28 (24–40)	0.174
Abdominal sepsis	141 (61%)	97 (54%)	44 (80%)	0.002
Bowel ischemia	67 (48%)	48 (27%)	19 (35%)	0.308
Hollow viscous perforation	51 (36%)	35 (20%)	16 (29%)	0.141
Inflammation/infection	20 (14%)	13 (7%)	7 (13%)	0.268
Anastomotic leak	3 (2%)	1 (1%)	2 (4%)	0.140
Trauma	79 (34%)	70 (39%)	9 (16%)	0.001
Injury severity score	33 (24–40)	33 (23–40)	40 (32–47)	0.062
Nontrauma hemorrhage/ACS	13 (6%)	11 (6%)	2 (4%)	0.738
On admission
Heart rate, beats/min	99 (84–113)	99 (83–113)	97 (84–109)	0.677
Systolic blood pressure, mm Hg	126 (106–143)	127 (107–145)	121 (105–138)	0.192
Mean arterial pressure, mm Hg	85 (73–98)	86 (74–99)	82 (71–98)	0.371
Vasopressor infusion	43 (19%)	33 (19%)	10 (18%)	>0.999
Lactate, mmol/L	2.5 (1.3–3.8)	2.5 (1.3–4.2)	2.4 (1.4–3.3)	0.733
Creatinine, mg/dL	1.0 (0.8–1.3)	1.0 (0.8–1.2)	1.0 (0.7–1.7))	0.578
International normalized ratio	1.3 (1.1–1.5]	1.3 (1.1–1.5)	1.3 (1.1–1.5)	0.556
White blood cells, ×109/L	12.3 (8.4–17.0)	12.7 (8.6–17.2)	11.3 (7.4–16.1)	0.131
Initial laparotomy
Hollow viscous resection or repair	143 (61%)	99 (56%)	35 (64%)	0.350
Solid organ resection or repair	64 (27%)	44 (25%)	6 (11%)	0.038
Washout/lysis of adhesions only	36 (15%)	20 (11%)	7 (13%)	0.810

Data are presented as median (interquartile range) or n (%).

ACS indicates abdominal compartment syndrome; ASA, American Society of Anesthesiologists; ICU, intensive care unit.

TABLE 2.

Management and Outcome Parameters

Management and Outcomes	All Patients (n = 233)	PFC (n = 178)	No PFC (n = 55)	p
Abdominal operations	2.0 (2.0–4.0)	2.0 (2.0–3.0)	4.0 (3.0–6.0)	<0.001
Interval between operations, d	1.3 (1.0–1.5)	1.3 (1.0–1.5)	1.4 (0.9–2.2)	0.696
Days to PFC	—	2.0 (1.0–3.0)	—	—
Intravenous fluid administration, L
Within 48 h of initial laparotomy	11.3 (8.5–14.5)	11.2 (8.5–13.9)	11.9 (7.9–17.2)	0.688
48–96 h after laparotomy	7.1 (4.7–9.6)	6.7 (4.3–9.3)	8.6 (6.1–12.9)	0.262
96 h to 7 d after laparotomy	6.9 (3.8–10.3)	6.2 (3.1–9.4)	8.7 (5.6–12.4)	0.775
NPWT output, L
Within 48 h of initial laparotomy		1.4 (1.0–2.5)	2.2 (1.6–3.4)	0.001
48–96 h after laparotomy		0.1 (0.0–1.4)	2.1 (1.0–3.5)	<0.001
96 h to 7 d after laparotomy		0.0 (0.0–0.5)	1.2 (0.5–3.9)	<0.001
Transfusions during admission
Red blood cell units	5.0 (1.0–10.5)	5.0 (1.0–9.0)	6.0 (1.0–16.0)	0.336
Plasma units	2.0 (0.0–7.0)	2.0 (0.0–6.0)	2.0 (0.0–8.0)	0.362
Hospital length of stay, d	22.0 (11.5–33.5)	20.0 (10.0–30.3)	27.0 (13.0–36.0)	0.066
ICU-free days	6.0 (3.0–12.0)	6.0 (3.0–11.0)	7.0 (1.0–16.0)	0.221
Days on mechanical ventilation	9.0 (4.0–17.0)	8.5 (3.0–16.3)	11.0 (6.0–18.0)	0.048
Inpatient mortality	27 (12%)	13 (7%)	13 (24%)	0.002

Data are presented as median (interquartile range) or n (%).

ICU indicates intensive care unit.

Serum Protein Levels

Patients who failed to achieve fascial closure had significantly higher CRP on admission (249 vs. 148 mg/L, p = 0.003), at the time of initial laparotomy (237 vs. 154, p = 0.002), and at discharge (124 vs. 72, p = 0.003) (Fig. 1). In the PFC group, serum albumin (in grams per deciliter) began to increase 48 hours after initial laparotomy. Among patients who failed to achieve fascial closure, albumin levels continued to decrease for 7 days following laparotomy and were significantly lower at 7 days (2.3 vs. 2.5, p = 0.028) and at discharge (2.5 vs. 2.8, p = 0.004). Prealbumin levels (in milligrams per deciliter) were similar between groups at each time point.

For each patient, nadir serum albumin (in grams per deciliter) was plotted against total milliliters of NPWT output during admission (Fig. 2). There was a significant inverse correlation (r = −0.33, p < 0.001) between these two parameters. There was a similar but weaker correlation between nadir prealbumin (in milligrams per deciliter) and NPWT output (r = −0.22, p = 0.001) and no significant correlation between peak CRP (in milligrams per liter) and NPWT output (r = 0.04, p = 0.615).

Figure 2.

Among patients who underwent exploratory laparotomy and TAC with an NPWT dressing, there was a significant correlation between nadir serum albumin and total NPWT fluid output during admission. Each patient is represented as a single bar.

Exogenous Albumin

Administration of exogenous albumin (expressed in grams per day) correlated with higher serum albumin levels (in grams per deciliter), and this association strengthened over time. The correlation between exogenous albumin (in grams per day) given within 48 hours following initial laparotomy and serum albumin at the 48-hour time point was 0.26 (p = 0.002). The correlation between exogenous albumin given from 48 hours to 96 hours and serum albumin at the 96-hour time point was 0.29 (p = 0.002). The correlation between exogenous albumin given from 96 hours to 7 days and serum albumin at 7 days was 0.40 (p < 0.001).

Predicting Primary Fascial Closure

Univariate associations between patient characteristics and failure to achieve PFC are listed in Table 3. Traumatic injury and the performance of a solid organ resection or repair at initial laparotomy were both significant on univariate analysis but were not included in the multivariable model because traumatic injury was essentially the inverse of abdominal sepsis (r = −0.879, p < 0.001), and solid organ resection or repair was a surrogate for traumatic injury (r = 0.657, p < 0.001). All other statistically significant variables were included in the multivariable model (Table 4). Abdominal sepsis was the strongest independent predictor of failure to achieve fascial closure (odds ratio [OR], 3.18 [95% CI, 1.52–6.64]), followed by serum albumin of less than 2.6 g/dL (OR, 2.59 [95% CI, 1.02–6.61]) and body mass index of greater than 40 kg/m² (OR, 2.42 [95% CI, 1.04–5.66]).When controlling for these factors, peak CRP of greater than 250 mg/L was not statistically significant (OR, 1.82 [95% CI, 0.93–3.57]). This model had an area under the receiver operating characteristic curve of 0.723 (0.646–0.801).

TABLE 3.

Univariate Associations Between Patient Characteristics and Failure to Achieve Primary Fascial Closure

Factors	OR	95% CI	p
Age >65 y	0.64	0.32–1.29	0.210
Body mass index* >40 kg/m2	2.70	1.26–5.79	0.011
Charlson Comorbidity Index >4	1.69	0.60–4.75	0.316
ASA physical status classification >3	0.97	0.51–1.87	0.936
Abdominal sepsis*	3.06	1.51–6.19	0.002
Trauma**	0.30	0.14–0.64	0.002
Nontrauma hemorrhage/ACS	0.57	0.12–2.67	0.478
On admission
Heart rate >110 beats/min	0.59	0.28–1.23	0.159
Systolic blood pressure <90 mm Hg	2.06	0.71–5.94	0.183
Mean arterial pressure <60 mm Hg	1.65	0.40–6.84	0.487
Vasopressor infusion	0.98	0.45–2.14	0.952
Lactate >3.5 mmol/L	0.57	0.25–1.30	0.181
Creatinine >2.0 mg/dL	2.20	0.93–5.63	0.071
International normalized ratio >1.5	0.91	0.42–1.98	0.808
White blood cells >18 × 109/L	0.77	0.36–1.66	0.499
Initial laparotomy
Solid organ resection or repair**	0.37	0.15–0.93	0.034
Hollow viscous resection or repair	1.40	0.75–2.61	0.294
Washout/lysis of adhesions only	1.15	0.46–2.89	0.763
Transfusions during admission
>5 Red blood cell transfusion	1.27	0.69–2.33	0.439
>3 Plasma transfusions	1.38	0.75–2.54	0.301
Nadir serum albumin* <2.6 g/dL	3.15	1.27–7.82	0.013
Peak CRP* >250 mg/L	2.15	1.16–3.97	0.015

All 233 patients in the study population were included in this analysis. Optimal cutoff values for continuous variables were assessed by Youden index.¹⁶

^*Included in multivariable model.

^**Excluded from multivariable model because of colinearity to other variables in the regression equation.

ACS indicates abdominal compartment syndrome; ASA, American Society of Anesthesiologists.

TABLE 4.

Multivariable Predictors of Failure to Achieve Primary Fascial Closure

Factors	OR	95% CI	p
Abdominal sepsis	3.18	1.52–6.64	0.002
Nadir serum albumin <2.6 g/dL	2.59	1.02–6.61	0.046
Body mass index >40 kg/m2	2.42	1.04–5.66	0.041
Peak CRP >250 mg/L	1.82	0.93–3.57	0.080

All 233 patients in the study population were included in this analysis. Area under the receiver operating characteristic curve for this model was 0.723 (0.646–0.801).

Exogenous albumin administration (in grams per day) was not suitable as a covariate because it was a surrogate for patients with more comorbidities and greater shock severity. Patients who received exogenous albumin were older (57 [47–71] vs. 53 [35–65] years, p = 0.010) and had higher Charlson Comorbidity Index scores (1.0 [0.0–3.0] vs. 0.0 [0.0–2.0], p = 0.001), lower admission systolic blood pressure (119 [102–138] vs. 133 [117–147] mm Hg, p < 0.001), lower admission mean arterial pressure (80 [70–99] vs. 90 [78–99] mm Hg, p = 0.008), and higher admission vasopressor infusion rates (25% vs. 9%, p = 0.003).

DISCUSSION

We observed that high levels of systemic inflammation and low levels of serum albumin were associated with failure to achieve PFC following TAC. Early and persistently elevated levels of CRP were associated with late hypoalbuminemia. These associations were independent of prealbumin levels, suggesting that changes in albumin levels were not surrogates for nutritional status. The significant correlation between hypoalbuminemia and NPWT output may represent a synergistic relationship in which systemic inflammation and capillary leak cause extravasation of albumin, creating a pathophysiologic oncotic gradient in the interstitial space. This increases ascites formation and NPWT protein losses, and the resulting hypoalbuminemia corrects slowly because of suppression of hepatic albumin synthesis. Albumin losses in ascites drained via NPWT may play a dominant role in the pathophysiology of hypoalbuminemia following TAC and failure to achieve PFC, given the significantly greater NPWT fluid losses among patients who did not achieve PFC. Notably, a significant difference in serum albumin levels between groups was not observed until 7 days after the initial laparotomy, after most PFCs had already occurred. Although late hypoalbuminemia may be a cause of failure to achieve fascial closure among patients who are not closed on their first or second relaparotomy, hypoalbuminemia may also be an effect of failure to achieve early fascial closure.

This hypothesis is supported by the observed correlations between exogenous administration and serum albumin levels over time. Administration of exogenous albumin correlated with higher serum albumin levels at each time point, but the weakest correlation was observed within 48 hours of laparotomy. This pattern may be clinically significant and appears to mimic burn pathophysiology. In burn patients, albumin administered within 12 hours of injury may extrude across permeable capillaries, pulling fluid into burned tissue and increasing lung water content, whereas albumin administered following resolution of capillary leak predominantly remains within blood vessels, drawing fluid into the intravascular space.^27–29 Although capillary leak is more difficult to describe among heterogeneous sepsis and trauma populations, the weight of evidence suggests similar underlying pathophysiology.^16–19 Briefly, pathogen-associated molecular patterns released from bacteria and damage-associated molecular patterns released from damaged cells activate Toll-like receptors, modulating the immune response and elaborating proinflammatory cytokines, which directly increase capillary permeability.^30–36 Reoperation for source control may also have influenced the results. Abdominal sepsis was the strongest predictor of failure to achieve fascial closure, and the observation that these patients had more abdominal operations than did other groups suggests that many of these patients had persistent inflammation due to residual infection requiring relaparotomy.

This study characterizes associations among systemic inflammation, serum albumin, NPWT fluid loss, and PFC for TAC patients but does not establish causal relationships. Given the expense of colloid fluids, their use cannot be recommended in the absence of randomized trials demonstrating that resuscitation with albumin is superior to crystalloid resuscitation alone for TAC patients. Unfortunately, such trials have not been reported to date. In a meta-analysis of 24 trials including a heterogeneous population of 9,920 critically ill patients, albumin did not improve survival compared with crystalloid resuscitation.³⁷ However, this study population does not accurately represent open-abdomen patients with NPWT dressings, and even in the absence of a mortality benefit, successful PFC may offset the costs of exogenous albumin. For TAC patients, current evidence supports physiologic resuscitation, early enteral nutrition, and treating sources of ongoing inflammation, each of which addresses potential causes and effects of hypoalbuminemia.^38–42

The major limitations of this study are its retrospective design, small sample size (n = 233), data from a single institution, and propensity to generate false-positive results by making multiple comparisons. Although single-institution studies inherently lack generalizability, this design allowed for standardization of resuscitation practices. Type I errors were avoided by performing analyses consistent with our purposes and hypotheses. In addition, CRP is a crude measure of systemic inflammation. Although it was not possible to perform a more detailed analysis of inflammatory biomarkers in this retrospective review, future studies would benefit from the availability of plasma and peritoneal fluid samples for this purpose. Also, patients with all forms of NPWT were included in this analysis. Animal^43,44 and human^45,46 studies suggest that different forms of NPWT have biologically and clinically important effects, and further research in the domain of TAC would benefit from a standardized approach NPWT. Future studies should also seek to determine if and when exogenous albumin administration produces a clinically significant benefit for TAC patients. Based on our data and data regarding burn injuries, it seems unlikely that early exogenous albumin is beneficial for patients with a systemic inflammatory response and capillary leak, but it remains possible that late exogenous albumin administration may be associated with modest benefits, albeit with considerable financial costs.

CONCLUSIONS

Early and persistent systemic inflammation was associated with late hypoalbuminemia and failure to achieve PFC. Serum albumin and NPWT output were inversely related. High NPWT may play an important role in the pathophysiology of hypoalbuminemia following TAC and failure to achieve PFC. Exogenous albumin administration was associated with increased serum albumin levels; this effect was weakest within 48 hours of laparotomy and strongest at 7 days after laparotomy, after most fascial closures had been achieved. High CRP and low serum albumin levels were independent predictors of failure to achieve PFC when controlling for the primary disease process, acute illness severity, and transfusion burden. Further investigation is required to determine whether late exogenous albumin administration offers a clinically significant benefit for TAC patients.