Association of Early Supplemental Parenteral Nutrition with Clinical Outcomes in Patients with Traumatic Brain Injury

Article information

J Korean Neurosurg Soc. 2026;69(5):770-778

Publication date (electronic) : 2026 July 16

doi : https://doi.org/10.3340/jkns.2026.0042

Heejin Kim¹, Jingyeong Kim¹, Jihyeon Ahn², Mihwa Jang³, Eunyoung Kim⁴, Dae-Sang Lee⁵, Min Ho Lee⁶, Hyung Sun Choi⁷, Maru Kim^,5

¹Department of Pharmacy, Uijeongbu St. Mary’s Hospital, Uijeongbu, Korea

²Nutrition Team, Uijeongbu St. Mary’s Hospital, Uijeongbu, Korea

³Nursing Team, Uijeongbu St. Mary’s Hospital, Uijeongbu, Korea

⁴Department of Surgery, Uijeongbu St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea

⁵Department of Trauma Surgery, Uijeongbu St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea

⁶Department of Neurosurgery, Uijeongbu St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea

⁷Department of Anesthesia and Critical Care, Section of Critical Care Medicine, University of Chicago, Chicago, IL, USA

Address for reprints : Maru Kim Department of Trauma Surgery, Uijeongbu St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 271 Cheonbo-ro, Uijeongbu 11765, Korea Tel : +82-31-820-5266, Fax : +82-31-820-5378, E-mail : maru@catholic.ac.kr

Received 2026 February 10; Revised 2026 March 9; Accepted 2026 March 24.

Abstract

INTRODUCTION

Several guidelines for critically ill patients have been reported. For nutritional support in the intensive care unit (ICU), recent guidelines from the European Society of Parenteral and Enteral Nutrition (ESPEN) recommend the early initiation of nutrition support [22]. Oral nutrition (ON) is the preferred option. ON is known to be physiological as it stimulates gastrointestinal motility, enzyme and hormone secretion [25]. It also preserves the pleasure of tasting and chewing, while enhancing patient autonomy and self-esteem. However, only approximately 30–60% of critically ill patients receive ON, and their nutritional intake is often insufficient [4,11]. If ON is not available, ESPEN recommends early enteral nutrition (EN) as the next option rather than early parenteral nutrition (PN) or delayed EN [22]. Early EN has been associated with reduced infectious complications compared to delayed EN, as well as shorter stays in the ICU and hospital. Additionally, EN helps maintain gut mucosal integrity [13].

However, some patients struggle to obtain sufficient nutrition through ON or EN alone. Patients with traumatic brain injury (TBI) often face challenges with oral diets or EN due to frequently diminished mental status. Neurologic deficits, such as a Glasgow coma score (GCS) of 8 or lower, are common indications for definitive airway management [16,23]. Gravesteijn et al. [8] reported that 40% of patients with TBI were initially intubated. For patients with endotracheal intubation, ON is limited. Even those who do not require intubation may experience dysphagia. Additionally, severely injured patients may have increased intracranial pressure, with nausea and vomiting being common clinical signs [18]. Patients with TBI, especially those with higher injury severity or lower levels of consciousness, are at risk for aspiration pneumonia [2]. As a result, it may not provide adequate nutrition through ON and EN to patients with TBI.

Current guidelines recommend providing 12–25 kcal/kg/day and 1.2–2.0 g/kg/day of protein in ICU patients [5]. Initial hypocaloric nutrition is recommended during the acute phase of illness [22]. After the 3rd hospital day, nutrition support can be gradually progressed to 80–100% of the target [24]. In particular, trauma patients showed higher energy and protein requirements [6,15]. Early adequate nutrition in the ICU is linked to improved survival rates in patients with TBI [17]. The survival group showed more calorie and protein supply in the analysis. However, oral diet or EN alone might be difficult to fulfil the target in these patients. A recent review article in trauma and critically ill patients suggests that a hybrid strategy of EN and PN may be ideal for meeting calorie and protein needs [7]. The role of supplementary PN (SPN) needs to be further delineated as robust evidence is lacking. Despite existing guidelines recommending early EN in critically ill patients, achieving adequate nutritional targets remains challenging in patients with TBI due to neurological impairment and intolerance to enteral feeding. While SPN has been proposed as a strategy to bridge this gap, evidence specifically evaluating its early use in patients with TBI is limited. Therefore, this study aimed to explore the association between early SPN and nutritional adequacy and clinical outcomes in patients with TBI. The authors hypothesized that early EN with SPN by hospital day 4 could provide adequate nutritional support and lead to better clinical outcomes. This study aimed to analyze the effect of SPN in patients with TBI.

MATERIALS AND METHODS

RESULTS

The characteristics of participants

A total of 260 adult patients with TBI who were admitted to the trauma ICU for at least 7 days were identified. After exclusions, 122 patients were included in the study, with 61 in the non-SPN group and 61 in the SPN group. The numbers of patients at each stage of inclusion and exclusion are summarized in Fig. 1.

Fig. 1.

Study flowchart. TBI : traumatic brain injury, TICU : trauma intensive care unit, EN : enteral nutrition, HD4 : hospital day 4, SPN : supplementary parenteral nutrition.

Characteristics of the non-SPN and SPN groups are summarized in Table 1. The non-SPN group was relatively younger, predominantly female, and had higher initial SBP and more cases of diabetes. In contrast, the SPN group exhibited greater injury severity in the face, chest, abdomen, extremities, and external areas. However, both groups had similar injury severity scores and RTSs. There was no patient who underwent deep sedation or TTM.

Table 1.

Baseline demographic and clinical characteristics of enrolled patients (n=122)

Comparison of nutritional support

Table 2 shows the amount of nutrition on the 4th day of hospitalization. More calories were provided through oral diet and EN in the non-SPN group than the SPN group. However, more calories (1508 vs. 852 kcal/day; p<0.001) and protein (79 vs. 41 g/day; p<0.001) were supplied in the SPN group. After adjusting for body weight, the SPN group showed more sufficient calorie (22.2 vs. 13.8 kcal/kg/day; p<0.001) and protein (1.17 vs. 0.66 g/kg/day; p<0.001) support. These values were closer to the nutritional targets suggested by current guidelines.

Table 2.

Amount of nutrition supply at the 4th day of hospitalization (n=122)

Comparison of clinical outcomes

The SPN group had fewer ICU admissions during hospitalization compared with the non-SPN group (1.10 vs. 1.46; p=0.003) (Table 3). The occurrence of unplanned ICU readmission was lower in the SPN group (19 [31.1%] vs. 6 [9.8%]; p=0.004). Additionally, the SPN group showed a significantly lower incidence of UTI compared with the non-SPN group (12 [19.7%] vs. 27 [44.3%]; p=0.004).

Table 3.

Analysis of clinical outcomes between SPN and Non-SPN group

Effect of SPN on clinical outcomes

Lastly, a logistic regression analysis was performed to verify risk factors for unplanned ICU readmission and UTI (Table 4). Liver cirrhosis was a risk factor for unplanned ICU readmission, and being female was a risk factor for UTI. Meanwhile, SPN was identified as a preventable factor for unplanned ICU readmission (B=-1.390; Exp[B], 0.249; (95% confidence interval [CI], 0.090–0.688; p=0.007) and UTI (B=-0.968; Exp[B], 0.380; 95% CI, 0.162–0.889; p=0.026), indicating potential clinical benefit in patients with TBI.

Table 4.

Results of logistic regression

Subgroup analyses

Subanalyses were conducted with stratification based on patient severity (Table 5). Subgroup analyses focused on patients with major TBI (GCS ≤8) and those with major trauma (ISS ≥15). In the analysis of patients with major TBI, a statistically significant lower rate of unplanned ICU readmissions was observed in the SPN group (8 [34.8%] vs. 1 [3.6%]; p=0.007). Additionally, in the analysis of patients with major trauma, a lower rate of UTI was associated with the SPN group (22 [44.0%] vs. 10 [20.0%]; p=0.010).

Table 5.

Results of subgroup analyses in patients with major TBI and major trauma

DISCUSSION

In this study, we aimed to assess the effects of early SPN in patients with TBI admitted to the ICU. Our findings demonstrated that SPN by the 4th day of hospitalization is associated with adequate nutritional support to meet the target (calorie : 90.0% vs. 55.2%; protein : 90.0% vs. 50.8%). Additionally, early SPN was linked to fewer UTIs and unplanned ICU readmissions, highlighting its benefits for patients in critical condition.

The higher prevalence of comorbidities in the non-SPN group may have contributed to the increased incidence of UTI. In particular, diabetes is a well-known risk factor for several infections [10]. To control potential confounding factors, multivariable logistic analysis was performed. Even after adjustment, SPN remained independently associated with a lower risk of UTI.

One of the main arguments against early SPN is based on the EPaNIC trial, which reported worse outcomes associated with early SPN, including longer hospital and ICU stays as well as increased infectious complications [3]. However, the primary participants in the trial were patients who underwent cardiac surgery, with a mean ICU stay of only 3 to 4 days. Less than 1% of these patients had neurological diseases. Therefore, the application of the EPaNIC trial results to nutrition support strategies for patients with TBI is limited. The CALORIES and NUTRIREA-2 trials reported no differences between early EN and early PN9,19). They imply that the route of nutritional support was not associated with infectious complications in cases of proper nutritional dosing. Additionally, the EAT-ICU trial reported no adverse outcomes related to SPN [1]. If nutritional support could not meet the target, the standard care group began SPN on the 7th day of hospitalization. Both the intervention and standard care groups exhibited similar mortality rates and clinical outcomes. However, patients with TBI were excluded from the EAT-ICU study. Recently, the NUTRIREA-3 trial highlighted potential drawbacks of early full nutrition [20]. The route of delivery of nutrition, be it EN, PN or SPN was not the primary focus of the trial. However, the enrolled patients primarily had respiratory and cardiac issues, while those with central nervous system problems or trauma constituted less than 10%. The median neurologic sub-score of the Sequential organ failure assessment score was only 1 point. By contrast, the unique merit of the present study is that it investigated the dual delivery of nutrition support in the ICU for patients with TBI. Enrolled patients had a mean GCS of 10 and a mean AIS of the head and neck of 4. The feasibility of SPN has not been sufficiently evaluated in patients who are unable to achieve adequate nutrition through oral diet or EN alone. This study may provide useful insights for developing nutritional support strategies in patients with TBI.

The addition of PN may increase the risk of overfeeding and should be approached with caution. In this study, five patients received more than 30 kcal/kg/day, all of whom were in the SPN group; pneumonia developed in four of these cases. Additionally, the risk of bloodstream infections must be considered. In patients receiving PN, whether through central or peripheral routes, intravascular catheterization is necessary and may elevate the risk of bloodstream infections. A meta-analysis has identified total PN as a significant risk factor for catheter-associated bloodstream infections [14].

Acute cholecystitis is one of the well-known complications in patients receiving PN. Long-term fasting and related cholestasis could increase its risk in patients undergoing long-term total parenteral nutrition [21]. However, this study analyzed the effects of early SPN, not total parenteral nutrition. All patients included in the study were receiving oral diet or EN on day 4 of hospitalization, and the risk of cholestasis was considered low. None of the patients developed acute cholecystitis during hospitalization.

In the present study, no significant difference in bloodstream infection rates was observed between the SPN and non-SPN groups. This finding may be attributed to the high baseline requirement for intravenous catheterization in critically ill patients. However, once patients with TBI are clinically stabilized and able to tolerate EN, further research is necessary to determine the need for SPN.

While excessive caloric intake through SPN should be avoided, additional research is needed to establish the optimal protein provision. A study evaluating urinary urea nitrogen in neurocritical patients undergoing TTM reported a mean protein requirement of 1.84 g/kg/day, which is higher than the 1.2–1.3 g/kg/day typically recommended for other critically ill populations [12]. Increased protein provision has also been recommended for critically ill trauma patients. Williams and Yeh [26] reported that PN delivering 1.5–2.5 g/kg/day of protein can be administered without a concomitant increase in other complications [7]. In the present study, all patients were individuals admitted to the ICU with TBI, and 77 of these patients had concomitant extracranial injuries. Critically ill trauma patients typically exhibit a hypermetabolic and inflammatory response, which may necessitate higher protein provision to meet their metabolic demands. However, further research is needed to establish appropriate protein intake targets for these patients.

CONCLUSION

In conclusion, this exploratory, hypothesis-generating study suggests that early SPN may be associated with improved nutritional adequacy and selected clinical outcomes in patients with TBI. These findings provide preliminary clinical evidence supporting the feasibility of early SPN in patients with TBI when EN alone is insufficient. Further well-designed studies on nutritional support in patients with TBI are needed.

Article information Continued