CLINICAL OUTCOMES OF INADEQUATE CALORIE DELIVERY AND PROTEIN DEFICIT IN SURGICAL INTENSIVE CARE PATIENTS. Nutrition in Critical Care

Transcription

1 Nutrition in Critical Care CLINICAL OUTCOMES OF INADEQUATE CALORIE DELIVERY AND PROTEIN DEFICIT IN SURGICAL INTENSIVE CARE PATIENTS By D. Dante Yeh, MD, Miroslav P. Peev, MD, Sadeq A. Quraishi, MD, MHA, MMSc, Polina Osler, MS, Yuchiao Chang, PhD, Erin Gillis Rando, RD, LDN, CNSC, Caitlin Albano, RD, LDN, CNSC, Sharon Darak, RD, LDN, CNSC, and George C. Velmahos, MD, PhD 6 American Association of Critical-Care Nurses doi: Background Adequate nutritional therapy in critically ill patients is integral to optimal outcome. Objective To evaluate the association between cumulative macronutrient deficit and overall morbidity in surgical intensive care unit patients. Methods Adult patients receiving enteral nutrition for more than hours were included if they had no previous admission to the surgical intensive care unit, had received no enteral feedings before admission, had no intestinal obstruction or ileus, and survived hours or more after admission. Data on demographics, outcomes, and nutritional intake during the unit stay were collected for up to days until oral intake began, discharge, or death. Outcome variables included lengths of stay in the hospital and intensive care unit, days with no mechanical ventilation, complications, and mortality. Results Of 9 participants, % were men, mean age was 6 years, and mean score on the Acute Physiology and Chronic Health Evaluation II was. Patients with high cumulative calorie deficit ( 6 cal) and high protein deficit ( g) had significantly fewer days with no mechanical ventilation (P ), longer unit stays (P ), longer hospital stays (P =.), more total complications (P =.), and more infectious complications (P =.9) than other participants. These associations remained significant in multivariable models after adjustments for age, sex, reason for admission, and propensity score of deficit. In-hospital and -day mortality did not differ. Conclusions Cumulative macronutrient deficits have important clinical outcomes in surgical intensive care patients. (American Journal of Critical Care. 6; :8-6) 8 AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 6, Volume, No.

2 Provision of adequate nutrition during critical illness is thought to be integral to achieving optimal health outcomes. - Providing timely, sufficient calories and protein is thought to influence both short-term outcomes (eg, intensive care unit [ICU] length of stay, ICU-acquired infections, duration of mechanical ventilation) and long-term outcomes (eg, hospital length of stay, discharge disposition). Observational studies, and randomized trials have indicated an inverse relationship between daily calories received and complication rates. In a large international observational study across 6 ICUs, Alberda et al found a stepwise decrease in mortality associated with each additional cal provided per day in underweight and overweight patients. Dvir et al also found a strong correlation between increases in energy deficit and increases in complications such as renal failure and sepsis. Accordingly, consensus statements - from professional nutrition societies emphasize initiating enteral nutrition within to 8 hours of ICU admission. Yet, despite strong recommendations and compelling supportive evidence, ICU patients receive only about one-half of prescribed nutrition in the first weeks of critical illness.,8 To date, medical ICU patients have been the focus of most research,9, on cumulative calorie or protein deficit. The evidence on the association between malnutrition and outcomes in surgical ICU patients is less convincing., Yet, compared with medical patients, surgical patients are more likely to have delayed initiation of enteral nutrition and to receive a lower percentage of prescribed calories. The purpose of our study was to investigate the association between calorie and protein deficits and important clinical outcomes in surgical ICU patients. About the Authors D. Dante Yeh is an assistant professor of surgery, Harvard Medical School, Boston, Massachusetts, and a staff surgeon and intensivist, Department of Surgery, Division of Trauma, Emergency Surgery and Surgical Critical Care, Massachusetts General Hospital, Boston, Massachusetts. Miroslav P. Peev is a general surgery resident, Tufts University, Boston, Massachusetts, and a research fellow, Department of Surgery, Division of Trauma, Emergency Surgery and Surgical Critical Care, Massachusetts General Hospital. Sadeq A. Quraishi is an assistant professor of anesthesia, Harvard Medical School, and a staff anesthetist and intensivist, Department of Anesthesia, Critical Care and Pain Medicine, Massachusetts General Hospital. Polina Osler is a medical student, Harvard Medical School. Yuchiao Chang is an assistant professor of medicine, Harvard Medical School, and a statistician, Department of Medicine, Division of General Internal Medicine, Massachusetts General Hospital. Erin Gillis Rando, Caitlin Albano,and Sharon Darak are critical care dietitians, Department of Nutrition and Food Services, Massachusetts General Hospital. George C. Velmahos is professor of surgery, Harvard Medical School, and division chief, Division of Trauma, Emergency Surgery and Surgical Critical Care, Massachusetts General Hospital. Corresponding author: D. Dante Yeh, MD, 6 Cambridge St, #8, Boston, MA ( dyeh@partners.org). Methods Patients and Setting We performed a prospective, observational, cohort study of patients from surgical ICUs in Massachusetts General Hospital, an academic hospital in Boston, Massachusetts. During the study period (March -December ), both surgical ICUs received patients from the trauma and emergency surgery service as well as from transplant, vascular, urologic, orthopedic, colorectal, and surgical oncology services. Surgical patients with medical (nonsurgery related) indications such as sepsis or rapid atrial fibrillation admitted to the ICUs were also considered for inclusion in the study. All patients 8 years and older who received enteral nutrition for more than hours were eligible. If parenteral nutrition or propofol was given concomitantly with enteral feedings, the amount of calories and protein content received from these sources was determined and included in the daily nutritional assessments. Goal rates for enteral feedings were adjusted to avoid hyperalimentation in patients receiving concomitant intravenous nutrition. For patients with multiple surgical ICU admissions, nutritional data were collected solely for the index ICU admission. Exclusion criteria were ICU stay less than hours, previous ICU stay within the same hospitalization, use of enteral feedings before ICU admission, and diagnosis of intestinal obstruction (mechanical or paralytic ileus). Because of the observational study design, Partners Human Research Committee (an institutional review board) waived the requirement to obtain informed consent. Adequate nutrition is important to good clinical outcomes. AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 6, Volume, No. 9

3 If the gastric residual volume was to ml, feeding was continued and promotility agents were started. Clinical Management Initiation and advancement of enteral nutritional intake were based on the official enteral feeding protocol approved by all intensivists in the surgical ICUs where the study was done. When no absolute contraindications were present, enteral feedings were started within 8 hours of ICU admission, initially at a rate of ml/h and increased by ml/h every hours until the desired goal was reached. The goal rate in all surgical ICU patients was calculated by qualified dietitians and was based on the American Society of Parenteral and Enteral Nutrition (ASPEN) guidelines for the provision and assessment of nutritional support therapy in adult critically ill patients. For patients with body mass index (BMI; calculated as the weight in kilograms divided by height in meters squared) less than, the goal was to kcal per kilogram of actual weight and. to g of protein per kilogram of actual weight. For BMI to, the goal was to kcal per kilogram of ideal weight or to kcal per kilogram of actual weight and more than g of protein per kilogram of ideal weight. For BMI greater than, the calorie goals were the same as those for as BMI to and more than. g of protein per kilogram of ideal weight. Enteral feedings were administered via nasogastric, postpyloric, or gastrostomy tubes. Feeding tolerance was evaluated by measuring gastric residual volumes (GRVs) every hours. If the GRV was less than ml, the tube feedings that were suctioned out for measurement were returned to the patient, and feeding was continued according to protocol. If the GRV was to ml, the residual was infused again, feeding was continued according to protocol, and use of promotility agents (metoclopramide and/or erythromycin) was started. If the GRV was greater than ml, the residual was infused again, enteral feeding was stopped, use of promotility agents was started, GRV was measured hours later, and if the GRV was less than ml, enteral feeding was resumed at the previous rate. Clinical Data Clinical data collected included age, sex, BMI, score on the Acute Physiology and Chronic Health Evaluation (APACHE) II, and admission diagnosis. Data on calorie and protein intake from enteral and parenteral sources were abstracted from the daily nursing flow sheet; collection began on the first day of ICU admission and continued for a maximum of days after admission unless oral intake began, the patient was discharged from the ICU, or death occurred. The data were collected by a research fellow through daily review of the electronic medical record and bedside paper flow sheets. When clarification was required, the clinical team was contacted. Outcomes were determined from review of hospital administrative data and discharge summaries. Calories contained in propofol and the use of supplements (eg, whey powder) were accounted for when total calorie and protein delivery were calculated. Adequacy of calories and protein received was expressed as a percentage of the prescribed nutrition. In our practice, assessments of resting energy expenditure, nitrogen balance, and serum levels of prealbumin and C-reactive protein are obtained solely on selected patients who require long-term critical care and therefore were not routinely measured in the patients in our study during the first weeks after surgical ICU admission. Outcomes The primary outcomes of interest were ICU length of stay, hospital length of stay, and 8-day ventilator-free days (VFD; number of days in a 8-day period that no mechanical ventilation was required). The secondary outcome of interest was the -day complication rate. Gastrointestinal complications were abdominal distention that required withholding enteral feedings, diarrhea that involved a documented workup and required cessation of enteral feedings, persistent nausea and vomiting, and clinically important aspiration events with radiological evidence and detection of enteral feeding material in tracheal aspirate (on visual inspection). Infectious complications included pneumonia (defined as radiological findings of new or progressive infiltrate, and at least of the following: fever [temperature > 8 C], leukocytosis or leukopenia, purulent secretions, or temporally relevant confirmed aspiration event); catheter-related infection, surgical site infections, urinary tract infection, and other nonspecified type of infection confirmed by microbiological culture data. Cardiovascular complications included acute myocardial infarction, new-onset atrial fibrillation, and new-onset congestive heart failure. Complications related to wound healing encompassed anastomotic leak, wound dehiscence, and new-onset pressure ulcers. Statistical Analysis Continuous variables were summarized by using means with standard deviations or medians with interquartile ranges (IQRs), whichever was more appropriate; categorical variables were summarized AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 6, Volume, No.

4 by using frequencies. On the basis of the existing literature on nutritional deficits and outcomes in critical illness, patients were dichotomized into low- and high-deficit groups. For cumulative calorie deficit, patients were dichotomized to less than 6 kcal and 6 or more kcal and for cumulative protein deficit, to less than g and g or greater. Outcomes between the groups were compared by using -sample t tests, Wilcoxon rank sum tests, or the Fisher exact test as appropriate. In an attempt to reduce confounding effects, we controlled for age, sex, BMI, and propensity score of nutritional deficits in all multivariable regression analyses. The propensity score of nutritional deficits was constructed by using logistic regression models that included APACHE II score, reason for surgical ICU admission, timing of initiation of enteral feeding, and presence or absence of gastrointestinal surgery. Specifically, to investigate the association between nutritional deficits and length of stay, we compared deficit groups by using a linear regression model for the log-transformed length-of-stay outcomes, because length of stay in the analytic cohort was not normally distributed. To investigate the association between nutritional deficits and VFD, we compared deficit groups by using a linear regression model, because VFD in the analytic cohort was almost normally distributed. To investigate the association between nutritional deficits and -day complication rate, we compared deficit groups by using a Poisson regression, because complications in the ICU are discrete-count data variables. As a sensitivity analysis, reason for admission was added to all the multivariable models. On the basis of preliminary data from our own patient cohorts, we assumed that 6% of patients in the low-deficit groups were at or below the median ICU length of stay, -day VFD, and hospital length of stay. We also assumed that % of patients in the high-deficit groups were at or below the median ICU length of stay, -day VFD, and hospital length of stay. To detect this difference with a power of 8% and assuming equals., we would require a minimum of patients in each group. All analyses were conducted by using SAS, version 9., software (SAS Institute). Two-sided P values less than. and 9% CIs that did not include were considered statistically significant. Results The analytic cohort consisted of 9 surgical ICU patients who met inclusion and exclusion criteria (Table ). The majority of the patients were male (%), and they had a mean age of 6 years (SD, years) and a mean APACHE II score of (SD, 6). A total of % of all patients received supplemental parenteral feedings, 8% received whey protein supplements, and 9% received calories from the administration of propofol. In more than two-thirds of the patients, enteral feedings were initiated within 8 hours of ICU admission (see Figure). Median duration of enteral feedings was days (IQR, 6- days). Overall, patients received 68% (SD, 9%) of their calorie and 6% (SD, %) of their protein requirements. The mean daily calorie deficit was kcal (SD, 6 kcal), and the mean daily protein deficit was g (SD, g). The mean total calorie deficit was 9 kcal (SD, 8 kcal), and mean total protein deficit was 98 g (SD, g). Median ICU length of stay in the analytic cohort was days (IQR, - days), and hospital length of stay was 9 days (IQR, 9- days). The median 8-day VFD was 9 (IQR, -). During hospitalization, a total of patients (6%) died. Seven of those deaths (%) occurred within days after admission to the surgical ICU. A total of complications occurred in the study cohort (Table ), with a median of (IQR, -) complications per patient. More than half of the complications were due to infections; pneumonia was a leading causative factor. Calorie Deficit Multivariable regression analysis (Table ) indicated that patients with a cumulative caloric deficit of 6 kcal or more had almost a -fold risk for prolonged ICU length of stay (risk ratio in the log scale,.; 9% CI,.6-.) and hospital length of stay (risk ratio in the log scale,.; 9% CI,.-.9) compared with patients with a cumulative calorie deficit of less than 6 kcal. Patients with a cumulative calorie deficit of 6 kcal or greater also had approximately fewer VFDs than did patients with a cumulative calorie deficit of 6 kcal or less ( = -6.9; 9% CI, -. to -.6). And finally, patients with a cumulative calorie deficit of 6 kcal or greater were more than. times more likely to experience complications during hospitalization (incident risk ratio,.66; 9% CI,.6-.) compared with patients with a cumulative caloric deficit of 6 kcal or less. No difference in in-hospital or -day mortality was detected between low- and high-calorie deficit groups. Protein Deficit Multivariable regression analysis (Table ) indicated that patients with a cumulative protein deficit of g or greater had almost a -fold risk of prolonged ICU length of stay (risk ratio in the log scale, Patients were divided into low- and high-deficit groups. AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 6, Volume, No.

5 Table Baseline demographics, nutrition prescription and delivery, and clinical outcomes a All Calorie deficit Protein deficit (N = 9) < 6 kcal (n = ) 6 kcal (n = ) P < g (n = ) g (n = ) P Age, mean (SD), y 6 () 6 (8) 9 (). 66 (6) 8 (). Male sex, No. (%) 6 () 6 (6) (9). (6) (89). Body mass index, b mean (SD) 6.6 (6.).6 (.6) 9.8 (6.6)..6 (.9) 8. (6.).6 APACHE II score, mean (SD). (6.). (6.6). (.).9. (6.). (6.6).9 Admission category, % Elective Emergency surgery Trauma Nonoperative (medical) Initiation of enteral nutrition, hours after admission, No. (%) < > 96 6 (6) () () () (6) 8 () 6 (9) () 8 (8) (9) () (9). (6) (9) () (9) 6 () () (8) (8). Nutrition, No. (%) Parenteral Protein supplements Propofol 9 () 6 (8) 8 (9) 6 (9) (8) 6 (9) () 6 (6) (96).68 >.99 >.99 () () (89) () 9 () 6 (9)..6. Baseline nutritional prescription, mean (SD) Calories, kcal/d Protein, g/d 8 (89) 8 () 6 () 8 () 989 () 9 (8). 89 () 8 () 89 (96) 98 () Daily nutrition, mean (SD) Calories, kcal/d Protein, g/d 8 (6) () () 9 () 9 (9) ().. (68) 8 () (6) ().9. ICU nutrition deficit, mean (SD) Caloric, kcal/d Protein, g/d (6) () 6 () () (8) () 86 () () () () Total calorie deficit, mean (SD), kcal 9 (8) 9 (8) () 668 () 86 (8) Total protein deficit, mean (SD), g 98 () () 9 () (8) () Percentage of calories delivered, mean (SD) 68 (9) (6) 9 (8) () 6 (). Percentage of protein delivered, mean (SD) 6 () (8) 8 (9) (8) 6 (). Days of enteral nutrition, median (IQR) (6-) 8 (-) (8-). (-) (8-) Ventilator days, median (IQR) (-) (-) (-9) (-) (-9) 8-day VFD, median (IQR) 9 (-) (6-) (-). (8-6) (6-9) ICU LOS, median (IQR), d (-) (8-) (6-) (8-6) (6-) Hospital LOS, median (IQR), d 9 (9-) (8-) 8 (-). (8-) 9 (9-) In-hospital mortality, No. (%) (6) () (). 9 (6) 6 (6) >.99 -day mortality, No. (%) () (6) ().6 () (8) >.99 Total complications, median (IQR) (-) (-) (-). (-) (-).6 Gastrointestinal complications, median (IQR) (-) (-). (-) (-).8 Infectious complications, median (IQR) (-) (-) (-).9 (-) (-).6 Cardiovascular complications, median (IQR) (-) (-). (-).6 Skin complications, median (IQR) (-).9 (-). Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation; ICU, intensive care unit; IQR, interquartile range; LOS, length of stay; VFD, ventilator-free days. a Because of rounding, not all percentages total. b Calculated as weight in kilograms divided by height in meters squared. AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 6, Volume, No.

6 .; 9% CI,.-.) and hospital length of stay (risk ratio in the log scale,.; 9% CI,.8-.) compared with patients with a cumulative protein deficit less than g. Patients with a cumulative protein deficit of g or greater also had approximately fewer VFDs compared with patients with a cumulative protein deficit less than g ( = -9.; 9% CI, -.66 to -6.9). And finally, patients with a cumulative protein deficit of g or more were approximately. times more likely to experience complications during hospitalization (incident risk ratio,.9; 9% CI,.-.) than were patients with a cumulative protein deficit less than g. No difference in hospital or -day mortality was detected between low- and high-protein deficit groups. Sensitivity Analysis Addition of reason for admission to the surgical ICU to the multivariable models to test the association of calorie and protein deficit with outcomes did not materially change the results of our primary analysis (data not shown). Moreover, because we observed a high correlation between cumulative calorie deficit and cumulative protein deficit (r =.8), we repeated each of our primary multivariable analyses for calorie deficit with protein deficit as an interaction term. Percentage of patients < > 96 Timing, hours Hours > < 8 Figure Initiation of enteral nutrition after admission to surgical intensive care unit. We found no interaction between calorie and protein deficit in any of the multivariable regression models (data not shown). Discussion In this prospective, cohort study of patients admitted to the surgical ICU, we found that cumulative Table Complications during hospital admission All Calorie deficit Protein deficit Complication (N = 9) < 6 kcal (n = ) 6 kcal (n = ) < g (n = ) g (n = ) Total number of complications 8 8 Number of complications per patient, median (IQR) (-) Gastrointestinal Abdominal distention Diarrhea Vomiting Aspiration events Infectious Pneumonia Aspiration pneumonia Catheter-related infection Surgical site infections Urinary tract infections Other Cardiovascular Myocardial infarction Atrial fibrillation Heart failure Skin and connective tissue Wound healing Wound dehiscence Pressure wounds AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 6, Volume, No.

7 Table Multivariable model estimates of calorie or protein deficit effect on outcomes Outcome Calorie deficit 6 kcal Protein deficit g Estimate 9% CI P Estimate 9% CI P Ventilator-free days, difference in days Length of stay in surgical intensive care unit, risk ratio in the log scale Length of stay in hospital, risk ratio in the log scale Total number of complications, incident risk ratio to to to to to.9.6 to to.. to.. calorie and protein deficit is associated with important clinical outcomes. Specifically, compared with the other patients, patients with an accumulated caloric deficit of 6 kcal or greater or those with an accumulated protein deficit of g or greater had longer ICU length of stay, longer hospital length of stay, fewer VFDs, and higher rate of complications. Although our results are compelling, a causal relationship cannot be inferred because of the observational nature of the study; nonetheless, our results provide evidence for a better understanding of how attention to nutritional status may promote desired outcomes in surgical ICU patients. For example, the concept of a cumulative nutritional deficit is analogous to using lactate as a biomarker for oxygen debt or cumulative fluid balance to account for crystalloid resuscitation. Serial measurements of lactate clearance are more prognostic of outcome than is initial serum level of lactate at the time of admission. Likewise, a patient with septic shock may require aggressive fluid replacement early in the hospital course, but later outcomes Patients with a cumulative caloric deficit of 6 kcal or more had longer ICU and hospital lengths of stay. will depend on how aggressively that accumulated fluid can be removed through diuresis after the patient s condition is stabilized. The same relationship may be true of nutritional deficits. Future investigations of calorie and protein deficit should elaborate on several questions: What is the interaction between time and calorie deficit? Is a calorie deficit of less than 6 calories that develops during a -day period as deleterious as one that develops rapidly in the first to days? If a causal link exists, can patients' outcomes be improved by paying down the deficit over time (ie, mild overfeeding to make up for prior deficits)? Villet et al previously reported that in surgical ICU patients, a cumulative deficit of less than kcal during the first week of care is associated with worse outcomes. Our threshold was much lower, and the deficit accumulated more insidiously during a -week period. In medical patients, Faisy and colleagues 9,6 found that cumulative deficits were predictive of ventilator-associated pneumonia caused by Staphylococcus aureus and suggested a daily energy deficit of kj (approximately 99 cal) as a threshold for predicting ICU mortality. In our study, the daily calorie deficit threshold for predicting complications and increased length of stay in surgical patients was remarkably similar ( cal); however, differences in mortality in our study were not significant, most likely because of the small number of patients. Plausibly, accumulating a macronutrient deficit would result in worse clinical outcomes (eg, increased rate of infections, wound dehiscence, decubitus ulcers, and longer stays) due to decreased immunocompetence, skeletal muscle catabolism, and impaired wound healing. Indeed, our data revealed an association between energy (calorie) deficit and prolonged mechanical ventilation, ICU length of stay, and complications. Other investigators have concluded opposite findings: that aggressive delivery of nutrients may be associated with harm.,8 However, accounting for confounding factors is important. First, patients who have short ICU stays and are generally healthy most likely will have good outcomes regardless of poor delivery of nutrients during the patients short stay. Second, days after permanent progression to oral intake may be misrepresented as % enteral intake because no tube feedings were delivered. Controlling for these confounding factors confirms our findings that improved nutrient delivery is correlated with improved clinical outcomes. 9 None of our patients had an absolute contraindication to enteral feedings, and all study patients received enteral feedings within the first weeks after ICU admission. Among patients with a calorie deficit greater than the 6-calorie threshold, 8% had enteral feedings within 8 hours of ICU admission. Interestingly, the groups with high calorie and protein deficits were prescribed significantly more calories and protein and had enteral feeding started on similar ICU days compared with groups with low calorie and protein deficits. Yet, patients in the groups with high calorie and protein deficits rapidly accumulated a deficit within a median of 6 days. This finding implies that prescription and initiation of enteral feeding were not the responsible AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 6, Volume, No.

8 culprits for deficit accumulation, rather the culprits were execution and delivery of feeding. Therefore, future efforts may benefit from a focus on optimizing the delivery of nutrients. For example, the rate of accumulation of the calorie and protein deficits was greatest early in the ICU stay. This result may have been due to our current policy of initiating enteral feeding at a low rate and subsequently titrating up the rate to ensure tolerance. This practice is based on local custom. In a randomized trial of head-injured patients, an aggressive strategy of initiating enteral feeding immediately at goal rate resulted in not only improved nutrient delivery but also fewer complications and infections. Conversely, in a study of mostly medical ICU patients with acute respiratory distress syndrome, patients randomized to either initial lowvolume (ie, trophic) enteral feeding or full- energy feeding for the first 6 days of mechanical ventilation did not differ in terms of VFD or all-cause mortality, despite clear separation of groups according to cumulative calorie deficits. However, neither group received the recommended. g/kg of protein, and this fact may have confounded the findings. Although Rice et al concluded that deliberate underfeeding of patients receiving mechanical ventilation is not demonstrably worse than early advancement to full-energy feeding, whether or not these results can be extrapolated to surgical patients is unclear. In some patients, the hourly rates of delivery of enteral feedings were deliberately left at a low, trophic rate at the request of the surgeon. Unfortunately, we did not specifically collect data on trophic tube feeding orders, and we can only infer this practice on the basis of the delivery of only a few calories, which persisted beyond what would be expected when titrating up the rate without evidence of interruption in enteral feeding. One common reason for nutrient inadequacy in the ICU is interruptions in continuous hourly tube feedings for intubation or extubation or procedures. In some instances, such as for procedures requiring conscious sedation or endotracheal intubation in patients without a controlled airway, these interruptions are unavoidable. Therefore, a strategy of -hour volume-based feeding may be used in lieu of the standard hourly ratebased method. This shift is feasible, safe, and efficacious in improving calorie delivery in ICU patients and is a key component of the PEP up (Enhanced Protein-Energy Provision via the Enteral Route) protocol described by Heyland and colleagues., Our results must be interpreted in the context of a number of limitations. The patients were recruited from a single, tertiary care facility, limiting the generalizability of our findings. The enteral feeding protocol we used is more aggressive than are other published feeding algorithms, although we still have room for improvement, especially for the policy of titrating up the rate and the selective (rather than mandatory) use of prokinetic medications. Because we focused primarily on a limited number of surgical subspecialties, the results may not be applicable to all surgical patients (eg, cardiac surgery or neurosurgery). Additionally, we did not attempt to characterize baseline nutritional status in our analytic cohort (expect for BMI), because existing biomarkers (eg, serum levels of albumin or prealbumin) and screening tools for malnutrition (eg, Malnutrition Screening Tool or Mini Nutritional Assessment-Short Form) are poorly reflective of nutritional status in critically ill patients.,6 Our calculation of nutritional deficit was based on daily calorie and protein requirements according to the guidelines of the American Society for Parenteral and Enteral Nutrition and was not based on indirect calorimetry or nitrogen balance studies, neither of which are routine practice for patients receiving enteral feeding in our surgical ICUs. Although we attempted to control for various clinically relevant variables that may affect the relationship between nutritional deficits and ICU outcomes, residual confounding might exist, for which we were unable to control. For example, intolerance of enteral feeding may be a marker of the severity of illness and the consequence, rather than the cause, of poor outcome. And finally, although the relatively small number of patients in our study may have been sufficient to address the primary study outcome, we were underpowered to detect differences in mortality, although the crude rate was higher in the higher deficit groups. As such, our findings should be considered useful in generating hypotheses to be tested in future randomized, controlled clinical trials. Despite these limitations, we think our results have value for bedside clinicians because we have defined a discrete threshold for cumulative calorie and protein deficit. When patients exceed this threshold, important clinical outcomes seem to be worse. The next logical step in investigation is to design an intervention to treat patients and prevent them from crossing these calorie- and protein-deficit thresholds and see if that intervention influences the clinical outcome. Conclusion In this prospective observational study, we found that both calorie and protein deficits may independently influence clinical outcomes in surgical ICU patients. Patients with higher nutritional Both calorie and protein deficits may affect outcomes in surgical intensive care unit patients. AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 6, Volume, No.

9 deficits were more likely to experience longer ICU length of stay, prolonged hospital length of stay, fewer VFDs, and higher complication rates than were patients with low calorie and protein deficits. Randomized, controlled trials are needed to determine whether minimizing cumulative nutritional deficits in surgical ICU patients improves clinical outcomes. ACKNOWLEDGMENTS This research was performed at Massachusetts General Hospital. Material in this article was presented as a poster at the Society of Critical Care Medicine rd Critical Congress; January 9-, ; San Francisco, California. FINANCIAL DISCLOSURES None reported. eletters Now that you ve read the article, create or contribute to an online discussion on this topic. Visit and click Submit a response in either the full-text or PDF view of the article. SEE ALSO For more about nutrition in critical care, visit the Critical Care Nurse website, and read the article by Mauldin and O Leary-Kelley, New Guidelines for Assessment of Malnutrition in Adults: Obese Critically Ill Patients (August ). REFERENCES. Rubinson L, Diette GB, Song X, Brower RG, Krishnan JA. Low caloric intake is associated with nosocomial bloodstream infections in patients in the medical intensive care unit. Crit Care Med. ;():-.. Alberda C, Gramlich L, Jones N, et al. The relationship between nutritional intake and clinical outcomes in critically ill patients: results of an international multicenter observational study [published correction appears in Intensive Care Med. 9 ;():8]. Intensive Care Med. 9;():8-.. Villet S, Chiolero RL, Bollmann MD, et al. Negative impact of hypocaloric feeding and energy balance on clinical outcome in ICU patients. Clin Nutr. ;():-9.. Dvir D, Cohen J, Singer P. Computerized energy balance and complications in critically ill patients: an observational study. Clin Nutr. 6;():-.. Martindale RG, McClave SA, Vanek VW, et al; American College of Critical Care Medicine; A.S.P.E.N. Board of Directors. Guidelines for the provision and assessment of nutrition support therapy in the adult critically ill patient: Society of Critical Care Medicine and American Society for Parenteral and Enteral Nutrition: executive summary. Crit Care Med. 9;(): Heyland DK, Dhaliwal R, Drover JW, Gramlich L, Dodek P; Canadian Critical Care Clinical Practice Guidelines Committee. Canadian clinical practice guidelines for nutrition support in mechanically ventilated, critically ill adult patients. JPEN J Parenter Enteral Nutr. ;():-.. Kreymann KG, Berger MM, Deutz NE, et al; DGEM (German Society for Nutritional Medicine); Ebner C, Hartl W, Heymann C, Spies C; ESPEN (European Society for Parenteral and Enteral Nutrition). ESPEN guidelines on enteral nutrition: intensive care. Clin Nutr. 6;():-. 8. Cahill NE, Dhaliwal R, Day AG, Jiang X, Heyland DK. Nutrition therapy in the critical care setting: what is best achievable practice? an international multicenter observational study. Crit Care Med. ;8(): Faisy C, Lerolle N, Dachraoui F, et al. Impact of energy deficit calculated by a predictive method on outcome in medical patients requiring prolonged acute mechanical ventilation. Br J Nutr. 9;():9-8.. Hoffer LJ, Bistrian BR. Why critically ill patients are protein deprived. JPEN J Parenter Enteral Nutr. ;():.. Engel JM, Muhling J, Junger A, Menges T, Karcher B, Hempelmann G. Enteral nutrition practice in a surgical intensive care unit: what proportion of energy expenditure is delivered enterally? Clin Nutr. ;():8-9.. Drover JW, Cahill NE, Kutsogiannis J, et al. Nutrition therapy for the critically ill surgical patient: we need to do better! JPEN J Parenter Enteral Nutr. ;(6):6-6.. Peev MP, Yeh DD, Quraishi SA, et al. Causes and consequences of interrupted enteral nutrition: a prospective observational study in critically ill surgical patients. JPEN J Parenter Enteral Nutr. ;9():-.. Zhang Z, Xu X. Lactate clearance is a useful biomarker for the prediction of all-cause mortality in critically ill patients: a systematic review and meta-analysis. Crit Care Med. ; (9):8-.. Wiedemann HP, Wheeler AP, Bernard GR, et al. Comparison of two fluid-management strategies in acute lung injury. N Engl J Med. 6;(): Faisy C, Candela Llerena M, Savalle M, Mainardi JL, Fagon JY. Early ICU energy deficit is a risk factor for Staphylococcus aureus ventilator-associated pneumonia. Chest. ;(): -6.. Krishnan JA, Parce PB, Martinez A, Diette GB, Brower RG. Caloric intake in medical ICU patients: consistency of care with guidelines and relationship to clinical outcomes. Chest. ; (): Arabi YM, Haddad SH, Tamim HM, et al. Near-target caloric intake in critically ill medical-surgical patients is associated with adverse outcomes. JPEN J Parenter Enteral Nutr. ; (): Heyland DK, Cahill N, Day AG. Optimal amount of calories for critically ill patients: depends on how you slice the cake! Crit Care Med. ;9(): Taylor SJ, Fettes SB, Jewkes C, Nelson RJ. Prospective, randomized, controlled trial to determine the effect of early enhanced enteral nutrition on clinical outcome in mechanically ventilated patients suffering head injury. Crit Care Med. 999;():-.. Rice TW, Mogan S, Hays MA, Bernard GR, Jensen GL, Wheeler AP. Randomized trial of initial trophic versus full-energy enteral nutrition in mechanically ventilated patients with acute respiratory failure. Crit Care Med. ; 9(): Heyland DK, Cahill NE, Dhaliwal R, et al. Enhanced proteinenergy provision via the enteral route in critically ill patients: a single center feasibility trial of the PEP up protocol. Crit Care. ;():R8.. Heyland DK, Murch L, Cahill N, et al. Enhanced protein-energy provision via the enteral route feeding protocol in critically ill patients: results of a cluster randomized trial. Crit Care Med. ;():-.. Barr J, Hecht M, Flavin KE, Khorana A, Gould MK. Outcomes in critically ill patients before and after the implementation of an evidence-based nutritional management protocol. Chest. ;():6-.. Jensen GL, Wheeler D. A new approach to defining and diagnosing malnutrition in adult critical illness. Curr Opin Crit Care. ;8(): Jensen GL, Bistrian B, Roubenoff R, Heimburger DC. Malnutrition syndromes: a conundrum vs continuum. JPEN J Parenter Enteral Nutr. 9;(6):-6. To purchase electronic or print reprints, contact American Association of Critical-Care Nurses, Columbia, Aliso Viejo, CA 966. Phone, (8) 899- or (99) 6- (ext ); fax, (99) 6-9; , reprints@aacn.org. 6 AJCC AMERICAN JOURNAL OF CRITICAL CARE, July 6, Volume, No.