Abstract
Objective
The prognostic nutritional index (PNI) has been reported as a significant predictor in various diseases. However, the prognostic value of the PNI in geriatric hip fracture patients has not been thoroughly evaluated. This study aimed to investigate the association between admission PNI and 3‐year mortality in those patients.
Methods
In this post hoc analysis, we included patients aged ≥65 years who underwent surgery for hip fracture between 2018 and 2019. The admission PNI was calculated as serum albumin (g/L) +5 × total lymphocyte count (×109/L). Patients were categorized into four groups based on PNI quartiles (≤ 43.55, 43.55–46.55, 46.55–49.20, and >49.20, respectively). The median follow‐up duration was 3.1 years. Cox proportional hazards models were used to calculate the hazard ratio (HR). Receiver operating characteristic curve (ROC) was conducted for using PNI to predict mortality.
Results
Of the 942 eligible patients, 190 (20.2%) patients died during the follow‐up. Compared to patients in the first quartile (Q1), those in the second (Q2), third (Q3), and fourth (Q4) quartiles had significantly lower mortality risks (HRs 0.50, 95% CI 0.35–0.74; 0.41, 95% CI 0.26–0.64; and 0.26, 95% CI 0.15–0.45, respectively). The optimal cutoff of PNI for predicting mortality was set as 45.275 (sensitivity, 0.674; specificity, 0.692; area under the curve (AUC), 0.727). Patients with higher PNI (>45.275) had a significant lower mortality risk (HR 0.39, 95% CI 0.28–0.55) compared to those with lower PNI (≤ 45.275).
Conclusion
PNI is a reliable and independent predictor of 3‐year mortality after hip fracture surgery in the elderly.
Keywords: Hip fracture, Mortality, Nutrition, Prognostic nutritional index
The prognostic nutritional index (PNI) is an objective and easy tool for nutritional assessment. Geriatric hip fracture patients with higher admission PNI values had significantly lower 3‐year mortality risk. An optimal PNI cutoff value for predicting mortality was set as 45.275.
Introduction
Hip fractures are common and serious injuries in elderly individuals, often caused by low‐energy trauma. 1 Hip fractures impose a heavy burden on patients and healthcare systems. 2 The World Health Organization predicts that the number of osteoporotic hip fractures will triple in the next 50 years, rising from 1.7 million in 1990 to 6.3 million in 2050 globally. 3 Studies have reported that hip fracture patients have a mortality rate of 20%–24% within 1 year, with ongoing risk of death for over 5 years. 4 Therefore, it is important to identify high‐risk patients before implementing interventions to improve outcomes.
Elderly patients with hip fractures often experience malnutrition. 5 Malnutrition has been identified as an independent risk factor for poor clinical outcomes in geriatric hip fracture patients following surgeries. 5 , 6 , 7 , 8 Consequently, it is imperative for surgeons to identify patients with malnutrition early in the hospital setting. Up to now, various nutritional screening tools (NSTs) have been proposed to evaluate patients' nutritional status, including subjective global assessment (SGA), 9 malnutrition universal screening tool (MUST), 10 and mini nutritional assessment (MNA). 11 However, NSTs are usually complex and some of them are subjective assessments which require skill and experience. An objective and easy tool for estimating nutritional status in geriatric hip fracture patients is needed.
The prognostic nutritional index (PNI), initially proposed by Buzby et al. 12 and refined by Onodera et al., 13 is a simple tool for evaluating perioperative nutrition of patients by using preoperative serum albumin (Alb) and total lymphocyte count (TLC). It was originally designed to assess patients' nutritional and immune status during gastrointestinal surgery. 12 To date, a low PNI has been identified as a significant predictor of poor outcomes and increased mortality in various diseases, including gastrointestinal cancer (PNI < 40), 13 colorectal cancer (PNI < 45.5), 14 and lung cancer (PNI < 45.5). 15 However, limited research has investigated the relationship between PNI and long‐term mortality risk in geriatric hip fracture patients, with varying and inconclusive findings. A study involving 263 elderly hip fracture patients found that those with a preoperative PNI over 38 had a lower risk of long‐term mortality, 16 but this result was not supported by another study after accounting for covariates. 17 Therefore, the objectives of this study are as follows: (i) to determine whether admission PNI is correlated with the 3‐year mortality risk in elderly patients undergoing surgery for hip fractures; and (ii) to determine the optimal cutoff value of PNI for predicting 3‐year mortality.
Materials and Methods
Study Design and Setting
The current study was conducted at a single tertiary hospital in Beijing, China, where a co‐managed orthogeriatric hip fracture care model was implemented. Ethics approvals were obtained from the Institutional Review Board at Peking University Health Science Centre (IRB00001052‐17021) and Biomedical Ethics Committee at Beijing Jishuitan Hospital (201807‐11) before starting the study. Written consent was obtained from all participants. For illiterate participants, informed consent to participate was obtained from their literate legal guardians. In this post hoc analysis, we utilized the baseline information gathered in advance from our prior observational research which focused on the effects of a co‐managed orthogeriatric hip fracture care model on older individuals with hip fractures in China (Clinical Trials.gov Identifier NCT03184896). 18 The research was conducted following the guidelines of the Helsinki Declaration.
Study Population, Recruitment and Follow‐up
Our study screened a total of 1092 patients aged 65 and older who had received surgery for hip fracture within 3 weeks after injury from November 2018 to November 2019. The clinical approach has been previously described. 18 During the screening, we excluded patients who met the following criteria: (i) those who did not have complete baseline or laboratory data at admission; (ii) those with pathological fractures or periprosthetic fractures; and (iii) those with terminal malignancies. We used the first blood test results taken at hospital admission to calculate PNI. Baseline serum Alb and TLC levels in peripheral blood were measured to calculate PNI using the formula: Alb (g/L) + 5 × TLC (109/L). According to the exclusion criteria, 150 patients were ruled out. A total of 942 eligible patients were divided into four groups based on PNI quartiles at admission (Q1 with PNI ≤43.55; Q2 with 43.45 < PNI ≤46.55; Q3 with 46.55 < PNI ≤49.20; Q4 with PNI >49.20). Patients were followed up by phone for a median duration of 3.1 (2.8, 3.4) years.
Data Collection
The current study prospectively gathered demographic and perioperative data. The demographic information encompassed age, sex, weight, height, body mass index (BMI), smoking and alcohol consumption habits, educational attainment, and residential situation. The educational status was classified into five levels, spanning from lacking literacy to having a university degree or higher. Baseline medical conditions such as hypertension, diabetes, cognitive and visual impairments were also documented. Cognitive function was evaluated with the mini‐mental state examination‐China (MMSE), and those with a score of 23 or below were classified as having cognitive impairment. 19 Cardiac function was evaluated by the left ventricular ejection fraction (LVEF) measured through preoperative echocardiography conducted in the emergency department. Upon admission, patients' depression severity was evaluated and quantified using the patient health questionnaire‐9 (PHQ‐9). 20 The Charlson comorbidity index (CCI) was computed as a measure of general health status.
Perioperative variables consisted of fracture pattern, American Society of Anesthesiologists (ASA) scores, anesthesia type, surgical procedure, time to surgery (TTS), and length of stay (LOS). At admission, results of laboratory blood examinations including blood routine and blood biochemical were collected. Nutritional status was indicated by serum Alb level and PNI. Additionally, the number of falls in the past year was documented, indicating the patients' tendency to be injured.
Undisplaced femoral neck fractures (FNFs) (Garden I/II), were treated with cannulated screws while displaced FNFs (Garden III/IV) were treated with total hip arthroplasty or hemiarthroplasty. Stable intertrochanteric fractures (ITFs) (31A1) were treated with dynamic hip screw (DHS), locking plate, or cephalomedullary devices, while unstable ITFs (31A2 to A3) and subtrochanteric fractures (STFs) were treated with only cephalomedullary devices. The operations were categorized as either internal fixation (including intramedullary nailing, DHS, locking plate, and cannulated screw fixation) or arthroplasty.
All patients had telephone follow‐ups by orthopedists 3 years after surgery. The follow‐up data included mortality and health‐related quality of life (HRQoL) which indicated by Euro‐Qol 5 dimensions score (EQ‐5D 5L). 21 The EQ‐5D instrument comprises a five‐level response scale (ranging from no problems to extreme problems) for five health domains pertaining to daily activities: mobility, self‐care, usual activities, pain and discomfort, and anxiety and depression. Subsequently, the responses were transformed into a composite score using a validated utility model specific to the Chinese population. 22
Study Outcomes
In the current study, the primary outcome was 3‐year all‐cause mortality. The secondary outcome was EQ‐5D utility at the last follow‐up.
Statistical Analysis
Continuous variables were presented as means and standard deviations (SD) or medians and interquartile ranges (IQR). Categorical data were displayed using frequencies and numerical distributions. The χ 2‐test was used to compare categorical variables between groups, whereas student's t‐test or Mann–Whitney U‐test were applied for continuous variables based on their distributions.
Covariates adjusted in multivariable models were determined based on baseline variables with a p‐value <0.05 from univariable analysis or those deemed clinically relevant. Hazard ratios (HRs) were calculated using multivariate Cox proportional‐hazards models to assess the relationship between PNI and mortality risk, with and without adjusting for covariates. Kaplan–Meier survival curves were conducted. A receiver operating characteristic curve (ROC) was also produced by using PNI to predict 3‐year mortality, with the optimal cutoff value determined by the Youden index. Restricted cubic spline (RCS) models were used to fit Cox proportional‐hazards models with the cutoff to observe the HR trend in relation to admission PNI. Additionally, multivariate linear regression models were utilized to evaluate the relationship between the admission PNI and the EQ‐5D utility value at the last follow‐up.
The statistical software R 4.1.1 1 (http://www.R-project.org, The R Foundation) was used for analysis, with p < 0.05 (two‐tailed) considered statistically significant.
Results
Population and Baseline Characteristics
Figure 1 shows the study's flowchart. The current study included a total of 942 eligible patients underwent surgery for hip fracture. As shown in Table 1, the mean age was 79.6 ± 7.8 years; 681 (72.3%) were women, and 440 (46.7%) were intertrochanteric fractures. Patients in Q1 (82.9 ± 7.6) were older than those in Q2 (80.1 ± 7.6), Q3 (78.6 ± 6.9), and Q4 (76.7 ± 7.8). The proportion of female was lower in Q1 (63.4%) compared to Q2 (70.6%), Q3 (75.2%), and Q4 (80.0%). Patients in Q1 had a lower value of BMI than those in Q2, Q3, and Q4. Less patients had diabetes at admission in Q1 (22.3%) compared to Q2 (25.1%), Q3 (32.1%), and Q4 (37.4%). Patients in Q1 had a higher proportion of anemia (68.9%, p < 0.001) and cognitive impairment (16.8%, p < 0.001) than those in Q2 to Q4. The PHQ‐9 score was higher in Q1 compared to Q2, Q3, and Q4 (p < 0.001). There were more patients with ITFs (60.1%) in Q1 compared to Q2 (45.5%), Q3 (45.7%), and Q4 (35.3%). Patients in Q1 had higher ASA scores, longer LOS, lower hemoglobin (Hb) and white blood cell (WBC) levels, higher serum creatinine, and lower Alb level. At last, 63 (6.7%) patients were lost during 3‐year follow‐up.
FIGURE 1.
Flowchart of the study. PNI, prognostic nutritional index; Q1, the first quartile; Q2, the second quartile; Q3, the third quartile; Q4, the fourth quartile.
TABLE 1.
Baseline characteristics.
| Variables | Total (n = 942) | Q1 (n = 238) | Q2 (n = 235) | Q3 (n = 234) | Q4 (n = 235) | p‐value |
|---|---|---|---|---|---|---|
| Age, years, Mean ± SD | 79.6 ± 7.8 | 82.9 ± 7.6 | 80.1 ± 7.6 | 78.6 ± 6.9 | 76.7 ± 7.8 | <0.001 |
| Female, n (%) | 681 (72.3) | 151 (63.4) | 166 (70.6) | 176 (75.2) | 188 (80.0) | 0.001 |
| BMI, kg/m2, Mean ± SD | 23.0 ± 3.9 | 21.8 ± 4.0 | 22.6 ± 4.1 | 23.4 ± 3.7 | 23.9 ± 3.3 | <0.001 |
| CCI, n (%) | 0.828 | |||||
| 0 | 313 (33.2) | 82 (34.5) | 73 (31.1) | 75 (32.1) | 83 (35.3) | |
| 1 | 339 (36.0) | 84 (35.3) | 87 (37.0) | 87 (37.2) | 81 (34.5) | |
| 2 | 190 (20.2) | 41 (17.2) | 49 (20.9) | 49 (20.9) | 51 (21.7) | |
| ≥3 | 100 (10.6) | 31 (13.0) | 26 (11.1) | 23 (9.83) | 20 (8.51) | |
| Comorbidities, n (%) | ||||||
| Diabetes, n (%) | 275 (29.2) | 53 (22.3) | 59 (25.1) | 75 (32.1) | 88 (37.4) | 0.001 |
| Hypertension, n (%) | 608 (64.5) | 146 (61.3) | 145 (61.7) | 160 (68.4) | 157 (66.8) | 0.27 |
| Anemia, n (%) | 495 (52.5) | 164 (68.9) | 139 (59.1) | 109 (46.6) | 83 (35.3) | <0.001 |
| Apoplexy, n (%) | 245 (26.0) | 59 (24.8) | 70 (29.8) | 58 (24.8) | 58 (24.7) | 0.508 |
| CAD, n (%) | 226 (24.0) | 50 (21.0) | 66 (28.1) | 60 (25.6) | 50 (21.3) | 0.202 |
| Cognitive impairment, n (%) | 102 (10.8) | 40 (16.8) | 31 (13.2) | 19 (8.12) | 12 (5.11) | <0.001 |
| Visual impairment, n (%) | 388 (41.2) | 93 (39.1) | 102 (43.4) | 97 (41.5) | 96 (40.9) | 0.818 |
| Ever or current smokers, n (%) | 158 (16.8) | 52 (21.8) | 37 (15.7) | 30 (12.8) | 39 (16.6) | 0.066 |
| Current drinkers, n (%) | 44 (4.7) | 11 (4.6) | 15 (6.4) | 7 (3.0) | 11 (4.7) | 0.387 |
| Educational level, n (%) | 0.13 | |||||
| Illiterate | 180 (19.1) | 53 (22.3) | 42 (17.9) | 47 (20.1) | 38 (16.2) | |
| Primary school or lower | 241 (25.6) | 74 (31.1) | 52 (22.1) | 59 (25.2) | 56 (23.8) | |
| High school | 253 (26.9) | 57 (23.9) | 64 (27.2) | 59 (25.2) | 73 (31.1) | |
| University or higher | 268 (28.5) | 54 (22.7) | 77 (32.8) | 69 (29.5) | 68 (28.9) | |
| MMSE scale, Mean ± SD | 20.5 ± 5.2 | 19.0 ± 6.0 | 20.3 ± 6.0 | 21.7 ± 3.6 | 21.1 ± 4.9 | <0.001 |
| PHQ‐9, Median (IQR) | 0.00 (0.00, 2.00) | 0.00 (0.00, 3.00) | 0.00 (0.00, 2.00) | 0.00 (0.00, 1.00) | 0.00 (0.00, 1.00) | <0.001 |
| Live alone, n (%) | 106 (11.3) | 27 (11.3) | 25 (10.6) | 28 (12.0) | 26 (11.1) | 0.975 |
| Live at home, n (%) | 564 (59.9) | 130 (54.6) | 135 (57.4) | 152 (65.0) | 147 (62.6) | 0.089 |
| Housebound, n (%) | 223 (23.7) | 78 (32.8) | 53 (22.6) | 48 (20.5) | 44 (18.7) | 0.001 |
| Unable to walk freely before fracture, n (%) | 275 (29.2) | 90 (37.8) | 65 (27.7) | 70 (29.9) | 50 (21.3) | 0.001 |
| Falling times in the last year, n (%) | 0.305 | |||||
| 0 | 383 (41.1) | 86 (36.9) | 96 (40.9) | 106 (45.9) | 95 (40.9) | |
| 1 | 423 (45.4) | 108 (46.4) | 104 (44.3) | 101 (43.7) | 110 (47.4) | |
| ≥2 | 125 (13.4) | 39 (16.7) | 35 (14.9) | 24 (10.4) | 27 (11.6) | |
| Non‐ground level fall, n (%) | 102 (10.9) | 26 (11.1) | 29 (12.3) | 24 (10.4) | 23 (9.91) | 0.849 |
| Fracture type, n (%) | <0.001 | |||||
| FNF | 483 (51.3) | 88 (37.0) | 124 (52.8) | 123 (52.6) | 148 (63.0) | |
| ITF | 440 (46.7) | 143 (60.1) | 107 (45.5) | 107 (45.7) | 83 (35.3) | |
| STF | 19 (2.02) | 7 (2.94) | 4 (1.70) | 4 (1.71) | 4 (1.70) | |
| Fracture side, n (%) | 0.581 | |||||
| Left | 441 (46.8) | 108 (45.4) | 113 (48.1) | 103 (44.0) | 117 (49.8) | |
| Right | 490 (52.0) | 125 (52.5) | 121 (51.5) | 129 (55.1) | 115 (48.9) | |
| Bilateral | 11 (1.17) | 5 (2.10) | 1 (0.43) | 2 (0.85) | 3 (1.28) | |
| LVEF, %, Median (IQR) | 65.4 ± 4.4 | 64.7 ± 5.2 | 65.7 ± 3.6 | 65.5 ± 4.6 | 65.8 ± 4.0 | 0.021 |
| Anesthesia type, n (%) | 0.497 | |||||
| Spinal | 940 (99.8) | 238 (100) | 234 (99.6) | 233 (99.6) | 235 (100) | |
| General | 2 (0.2) | 0 (0.00) | 1 (0.4) | 1 (0.4) | 0 (0.00) | |
| ASA score, n (%) | 0.001 | |||||
| 1 | 115 (12.2) | 26 (10.9) | 27 (11.5) | 25 (10.7) | 37 (15.7) | |
| 2 | 472 (50.1) | 96 (40.3) | 123 (52.3) | 121 (51.7) | 132 (56.2) | |
| ≥3 | 355 (37.7) | 116 (48.7) | 85 (36.2) | 88 (37.6) | 66 (28.1) | |
| TTS, hours, Median (IQR) | 26.7 (9.28, 47.2) | 26.4 (19.1, 49.0) | 28.0 (9.19, 47.2) | 26.0 (7.27, 46.5) | 26.8 (17.1, 46.7) | 0.416 |
| TTS within 48h, n (%) | 724 (76.9) | 172 (72.3) | 182 (77.4) | 183 (78.2) | 187 (79.6) | 0.253 |
| Operation type, n (%) | 0.003 | |||||
| Internal fixation | 551 (58.5) | 163 (68.5) | 132 (56.2) | 131 (56.0) | 125 (53.2) | |
| Arthroplasty | 391 (41.5) | 75 (31.5) | 103 (43.8) | 103 (44.0) | 110 (46.8) | |
| LOS, days, Median (IQR) | 5.00 (4.00, 6.42) | 5.54 (4.31, 6.96) | 4.96 (3.94, 6.46) | 4.94 (3.96, 6.04) | 4.92 (3.98, 6.02) | 0.002 |
| Laboratory test at admission | ||||||
| Hb, g/L, Mean ± SD | 115 ± 17.9 | 106 ± 16.5 | 115 ± 17.3 | 118 ± 17.2 | 123 ± 16.2 | <0.001 |
| WBC, ×109/L, Mean ± SD | 9.1 ± 2.8 | 8.2 ± 2.7 | 8.9 ± 2.9 | 9.6 ± 2.5 | 9.7 ± 2.9 | <0.001 |
| Platelet, ×109/L, Mean ± SD | 201 ± 73.3 | 203 ± 82.9 | 194 ± 68.6 | 200 ± 61.4 | 205 ± 78.4 | 0.387 |
| Crea, umol/L, Median (IQR) | 60.0 (49.0, 77.0) | 65.0 (50.0, 83.0) | 61.0 (50.0, 73.0) | 59.0 (50.0, 77.0) | 56.0 (47.0, 70.5) | 0.001 |
| Alb, g/L, Mean ± SD | 40.4 ± 3.5 | 36.5 ± 2.6 | 39.9 ± 1.9 | 41.7 ± 2.1 | 43.6 ± 2.9 | <0.001 |
| PNI, Mean ± SD | 47.5 ± 19.6 | 40.7 ± 2.4 | 45.2 ± 0.9 | 47.9 ± 0.8 | 56.2 ± 37.4 | <0.001 |
Abbreviations: Alb, (serum) albumin; ASA, American Society of Anesthesiologists; BMI, body mass index; CAD, coronary artery disease; CCI, Charlson comorbidity index; Crea, creatinine; FNF, femoral neck fracture; Hb, hemoglobin; IQR, interquartile range; ITF, intertrochanteric fracture; LOS, length of stay; LVEF, left ventricular ejection fraction; MMSE, mini‐mental state examination; PHQ‐9, patient health questionnaire‐9; PNI, prognostic nutrition index; Q1, the first quartile; Q2, the second quartile; Q3, the third quartile; Q4, the fourth quartile; STF, subtrochanteric fracture; TTS, time to surgery; WBC, white blood cell.
Primary Outcome
At the end of the follow‐up, 190 (20.2%) patients had died and mortality rates were significantly different between the four groups (42.0% [100/238] in Q1; 18.3% [43/235] in Q2; 12.8% [30/234] in Q3; 7.2% [17/235] in Q4; respectively) (Figure 2A). Kaplan–Meier curves demonstrated a lower survival probability in Q1 compared to Q2, Q3, and Q4 (p < 0.001) (Figure 2B). In the unadjusted model, patients in Q2 exhibited a significantly reduced mortality risk compared to those in Q1 (HR 0.38, 95% CI 0.27–0.54, p < 0.001). This trend was similarly observed in Q3 (HR 0.28, 95% CI 0.19–0.43, p < 0.001) and Q4 (HR 0.15, 95% CI 0.09–0.25, p < 0.001) (Table 2).
FIGURE 2.
Comparison of 3‐year mortality and Kaplan–Meier curves of four groups. (A) comparison of mortality. (B) Kaplan–Meier survival curves. PNI, prognostic nutritional index; Q1, the first quartile; Q2, the second quartile; Q3, the third quartile; Q4, the fourth quartile. *p < 0.05; ***p < 0.001.
TABLE 2.
HRs of admission PNI for 3‐year mortality.
| Admission PNI | Events, a n (%) | Unadjusted | Model 1 c | Model 2 d | ||||
|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | p‐value | HR (95% CI) | p‐value | HR (95% CI) | p‐value | |||
| Continuous | Per 1 | 190 (21.6) | 0.998 (1.002–0.986) | 0.757 | 1.002 (0.995–1.008) | 0.661 | 1.002 (0.996–1.008) | 0.464 |
| Quartiles | Q1 (≤43.55) | 100 (43.9) | 1 (reference) | 1 (reference) | 1 (reference) | |||
| Q2 (43.55–46.55) | 43 (19.3) | 0.38 (0.27–0.54) | <0.001 | 0.47 (0.32–0.67) | <0.001 | 0.50 (0.35–0.74) | <0.001 | |
| Q3 (46.55–49.20) | 30 (14.3) | 0.28 (0.19–0.43) | <0.001 | 0.39 (0.26–0.60) | <0.001 | 0.41 (0.26–0.64) | <0.001 | |
| Q4 (>49.20) | 17 (7.8) | 0.15 (0.09–0.25) | <0.001 | 0.22 (0.13–0.38) | <0.001 | 0.26 (0.15–0.45) | <0.001 | |
| Dichotomy b | Lower (≤45.275) | 128 (67.3) | 1 (reference) | 1 (reference) | 1 (reference) | |||
| Higher (>45.275) | 62 (32.6) | 0.27 (0.20–0.36) | <0.001 | 0.35 (0.26–0.49) | <0.001 | 0.39 (0.28–0.55) | <0.001 | |
Abbreviations: ASA, American Society of Anesthesiologists; BMI, body mass index; CI, confidence interval; Crea (serum), creatinine; Hb, hemoglobin; HR, hazard ratio; LOS, length of stay; LVEF, left ventricular ejection fraction; PHQ‐9, patient health questionnaire‐9; PNI, prognostic nutritional index; Q1, the first quartile; Q2, the second quartile; Q3, the third quartile; Q4, the fourth quartile; WBC, white blood cell.
Patients who were lost to follow‐up were excluded from the analysis,
Stratified by the optimal cutoff value of PNI that calculated by the Youden index,
Model 1, adjusted for age, sex, and BMI, and
Model 2, adjusted for age, sex, BMI, diabetes, cognitive impairment, PHQ‐9, housebound, unable to walk freely before fracture, fracture type, LVEF, ASA score, operation type, LOS, Hb, WBC, and Crea.
After adjusting for age, sex, and BMI in Model 1, the risk of mortality was significantly lower in Q4 (HR 0.22, 95% CI 0.13–0.38, p < 0.001), Q3 (HR 0.39, 95% CI 0.26–0.60, p < 0.001), and Q2 (HR 0.47, 95% CI 0.32–0.67, p < 0.001) compared to Q1. Subsequent adjustment for all covariates in Model 2 indicated that the mortality risk remained significantly lower in Q4 (HR 0.26, 95% CI 0.15–0.45, p < 0.001), Q3 (HR 0.41, 95% CI 0.26–0.64, p < 0.001), and Q2 (HR 0.50, 95% CI 0.35–0.74, p < 0.001) compared to Q1 (Table 2). The RCS models further demonstrated that an increase in admission PNI was associated with a continuous decline in mortality risk (Figure 3A,B).
FIGURE 3.
Association between admission PNI and 3‐year all‐cause mortality using restricted cubic spline (RCS) models. (A) RCS of PNI for 3‐year mortality without adjustment. (B) RCS of PNI for 3‐year mortality with adjusting for age, sex, BMI, diabetes, cognitive impairment, PHQ‐9, housebound, unable to walk freely before fracture, fracture type, LVEF, ASA score, operation type, LOS, Hb, WBC, and Crea. Patients who were lost to follow‐up were excluded from the analysis. ASA, American Society of Anesthesiologists; BMI, body mass index; Crea (serum), creatinine; Hb, hemoglobin; HR, hazard ratio; LOS, length of stay; LVEF, left ventricular ejection fraction; PHQ‐9, patient health questionnaire‐9; PNI, prognostic nutritional index; RCS, restricted cubic spline; WBC, white blood cell.
ROC analysis of the PNI identified an optimal cutoff value of 45.275 for predicting 3‐year mortality, with a sensitivity of 0.674, specificity of 0.692, and an area under the curve (AUC) of 0.727 (95% CI 0.686–0.768, p < 0.001) (Figure 4A). The Kaplan–Meier survival analysis demonstrated a significantly lower survival probability in patients with a PNI of 45.275 or lower (Lower group) compared to those with a PNI above 45.275 (Higher group) (p < 0.001) (Figure 4B). After adjusting for all covariates in Model 2, the Higher group exhibited a significantly reduced mortality risk compared to the Lower group (HR 0.39, 95% CI 0.28–0.55, p < 0.001) (Table 2).
FIGURE 4.
ROC of PNI for predicting 3‐year mortality and Kaplan–Meier survival curves of groups stratified by optimal cutoff value of PNI. (A) ROC of PNI for predicting 3‐year mortality (*Patients who were lost to follow‐up were excluded from the analysis). (B) Kaplan–Meier survival curves. The black dot pointed by the red arrow in the (A) represents the optimal cutoff value of PNI (45.275) for predicting 3‐year mortality (specificity 0.692, sensitivity 0.674). AUC, area under the curve; PNI, prognostic nutritional index; ROC, receiver operating characteristic curve.
Secondary Outcome
Higher admission PNI was initially linked to higher EQ‐5D in terms of HRQoL, but this association was no longer significant after adjustments (Table 3).
TABLE 3.
Association between admission PNI and EQ‐5D utility of the last follow‐up. a
| Admission PNI | Unadjusted | Model 1 c | Model 2 d | ||||
|---|---|---|---|---|---|---|---|
| B (95% CI) | p‐value | B (95% CI) | p‐value | B (95% CI) | p‐value | ||
| Continuous | Per 1 | 0.0005 (−0.0005–0.0015) | 0.605 | −0.0002 (−0.0011–0.0007) | 0.814 | −0.0009 (−0.0018–0) | 0.300 |
| Quartiles | Q1 (≤43.55) | 1 (reference) | 1 (reference) | 1 (reference) | |||
| Q2 (43.55–46.55) | 0.0760 (0.0359–0.1160) | 0.058 | 0.0401 (0.0018–0.0783) | 0.296 | 0.0014 (−0.0356–0.0384) | 0.970 | |
| Q3 (46.55–49.20) | 0.0809 (0.0407–0.1211) | 0.044 | 0.0300 (−0.0088–0.0688) | 0.440 | 0.0048 (−0.0333–0.0428) | 0.900 | |
| Q4 (>49.20) | 0.1665 (0.1273–0.2058) | <0.001 | 0.0926 (0.0537–0.1316) | 0.018 | 0.0461 (0.0063–0.0859) | 0.248 | |
| Dichotomy b | Lower (≤45.275) | 1 (reference) | 1 (reference) | 1 (reference) | |||
| Higher (>45.275) | 0.0958 (0.0669–0.1247) | <0.001 | 0.0567 (0.0286–0.0848) | 0.044 | 0.0282 (0.0001–0.0562) | 0.316 | |
Abbreviations: ASA, American Society of Anesthesiologists; BMI, body mass index; CI, confidence interval; Crea (serum), creatinine; EQ‐5D, Euro‐Qol 5 dimensions score; Hb, hemoglobin; LOS, length of stay; LVEF, left ventricular ejection fraction; PHQ‐9, patient health questionnaire‐9; PNI, prognostic nutritional index; PNI, prognostic nutritional index; Q1, the first quartile; Q2, the second quartile; Q3, the third quartile; Q4, the fourth quartile; WBC, white blood cell.
Only included patients who had completed 3‐year EQ‐5D follow‐up (n = 714),
Stratified by the optimal cutoff value of PNI that calculated by the Youden index,
Model 1, adjusted for age, sex, and BMI, and
Model 2, adjusted for age, sex, BMI, diabetes, cognitive impairment, PHQ‐9, housebound, unable to walk freely before fracture, fracture type, LVEF, ASA score, operation type, LOS, Hb, WBC, and Crea.
Discussion
This study shows strong evidence that the admission PNI reliably predicts 3‐year all‐cause mortality in elderly patients undergoing hip fracture surgery. Patients with higher PNI values showed a significantly lower risk of 3‐year mortality. The optimal PNI cutoff for predicting 3‐year mortality was identified as 45.275. Patients with PNI >45.275 had a 61% lower mortality risk compared to those with PNI ≤45.275. These findings highlight a clear dose–response relationship between the level of PNI and mortality, indicating that PNI status could serve as a potential maker for identifying geriatric patients at high risk of poor outcome following hip fracture surgery.
PNI as an Easy and Objective NST
Elderly patients with hip fracture often experience malnutrition which has been found to be associated with poor outcomes. 5 , 8 , 23 In the hospital setting, malnutrition is very common among older adults, with around half of them affected. 24 It is important to identify patients with malnutrition in hospital setting as early as possible for optimal nutritional care. 25 , 26 Up to now, many NSTs have been developed to evaluate nutritional status, including SGA, MUST, and MNA. However, NSTs are usually complex and some of them are subjective. In contrast, the PNI which is based on the results of a peripheral blood test allows a surgeon to easily and objectively assess the immune‐nutritional status of geriatric patients with hip fracture.
PNI Is an Independent Predictor for 3‐Year Mortality
Previous studies have described that the PNI has potential prognostic value in various populations, including patients with cancer, 27 , 28 , 29 and heart failure. 30 However, the prognostic value of PNI in geriatric patients with hip fractures has not been extensively investigated. A recent study which involved 221 hip fracture patients aged over 70 who underwent surgery revealed that patients with a higher PNI had a lower risk of all‐cause mortality at 5 years (HR 0.917, 95% CI 0.845–0.996, p = 0.039) compared to those with a lower PNI, 31 but this finding was not confirmed by another study after adjusting for potential confounders (HR 0.20, 95% CI 0.028–0.650, p = 0.067). 17
In this study, we found PNI is an independent prognostic factor for long‐term mortality that geriatric patients with a higher PNI had a significantly lower risk of 3‐year mortality after surgery. In the present study, we used a large prospective cohort of 942 geriatric hip fracture patients to evaluate the association between PNI and postoperative mortality. We included a wide range of potential confounders in our study. The study which found the PNI was not associated with mortality in hip fracture patients only included 80 patients and it was a retrospective study. 17 The limited sample size of the study may have limited the study's statistical power. Moreover, RCS models showed that higher admission PNI consistently reduced mortality risk, highlighting the reliability of PNI in predicting postoperative mortality in geriatric hip fracture patients.
The Optimal Cutoff Value of PNI
A cutoff value of 40 for the PNI has been proposed for perioperative assessment of gastrointestinal tumors. 13 However, there is still a lack of studies investigating the cutoff value for the PNI in predicting postoperative mortality among geriatric hip fracture patients after surgery. Recently, an observational study involving 263 hip fracture patients compared the risk of long‐term mortality between those with higher and lower PNI and they stratified patients by the PNI value of 38. 16 Their study indicated that patients with a preoperative PNI >38 have a reduced risk of long‐term mortality (HR 0.269, 95% CI 0.085–0.859, p = 0.027). However, the cutoff value they chose was indicated by another study which investigated the prognostic value of the PNI in patients with heart failure, rather than hip fracture. 32
A large cohort study which included 3351 patients (aged >45 years) with hip fractures underwent surgeries found that the patients in the medium‐category (43.23 < PNI ≤47.35) and high‐category (PNI >47.35) had lower hazard of 2‐year mortality (HRs 0.66, 95% CI 0.49–0.88; and 0.61, 95% CI 0.42–0.88, respectively), but they failed to find an optimal cutoff value for PNI to predict postoperative mortality which restricted its clinical application. 33 In this study, the cutoff value for PNI for the prediction of 3‐year mortality was set at 45.275. And patients with higher PNI (>45.275) showed a 61% decrease of 3‐year mortality risk compared to those with lower PNI (≤ 45.275). It would help surgeons to identify malnutrition easily and quickly in geriatric hip fracture patients with the application of the cutoff value.
Potential Mechanism
The mechanism underlying the association between PNI and long‐term mortality remains unclear. The PNI reflects both the nutritional and immunological status of patients with hip fracture. Therefore, nutritional deficiencies and compromised immune status may be the primary factors contributing to poor outcome. 5 , 34 However, further investigation is needed to clarify the potential mechanism between PNI level and mortality in geriatric patients.
Study Strengths and Limitations
The prospective data collection was a strength of the study, as it helped reduce recall bias and ensure internal validity during data collection. Furthermore, a substantial group of 942 elderly individuals with hip fractures and high follow‐up rates guaranteed a large enough sample size, aiding in the precise and dependable assessment of the relationship between PNI and postoperative mortality. This study has various limitations. Firstly, as a single center cohort study, the generalizability of the research results may be limited. Multicenter research would help in confirming our findings. Secondly, despite gathering data prospectively, participants were required to recollect and share information regarding the research findings during their scheduled telephone follow‐ups. This may have led to recall bias. Thirdly, despite attempts to adjust for all potential variables in the analysis, there may still be unidentified confounders that were not included in the study.
Conclusion
In conclusion, low admission PNI is a strong predictor for 3‐year mortality after hip fracture surgery in the elderly. Orthopedists may provide more nutritional support for patients with PNI under 45.275 to enhance their long‐term prognosis.
Conflict of Interest Statement
All authors declare that they have no conflict of interest.
Ethics Statement
Ethics approvals were received from the Institutional Review Board at Peking University Health Science Centre (IRB00001052‐17021) and Biomedical Ethics Committee at Beijing Jishuitan Hospital (201807‐11). All procedures used adhere to the tenets of the Declaration of Helsinki. All participants provided written informed consent before data collection. For illiterate participants, informed consent to participate was obtained from their literate legal guardians.
Author Contributions
All Authors contributed to the study conception, design and data interpretation. GL, MY, and XW had full access to the data, take responsibility for the content, and guarantee the integrity and accuracy of the work undertaken. YC, YG, ZT, WP, and JZ performed the data collection and analysis. YC, FG, CT, and GL did the literature search. YC, YG and MT did the measurements. YC and GL wrote the manuscript and review editing were performed by YM, JZ, MY, and XW. All authors have read and agreed to the published version of the manuscript.
Funding Information
This work was supported by Capital's Funds for Health Improvement and Research (code: 2018‐1‐2071, 2022‐1‐2071), Beijing Scholar Training Program 2021, and Beijing JST Research Funding (code: YGQ‐202214), the National Natural Science Foundation of China (No. 82372386).
Yimin Chen and Gang Liu contributed equally to this study.
Contributor Information
Minghui Yang, Email: doctyang0125@126.com.
Xinbao Wu, Email: wuxinbao_jst@126.com.
Data Availability Statement
The dataset is managed by Beijing Jishuitan Hospital. To request data access please contact the corresponding author.
References
- 1. Zhang Y‐W, Lu P‐P, Li Y‐J, Dai G‐C, Chen M‐H, Zhao Y‐K, et al. Prevalence, characteristics, and associated risk factors of the elderly with hip fractures: a cross‐sectional analysis of NHANES 2005‐2010. Clin Interv Aging. 2021;16:177–185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Papadimitriou N, Tsilidis KK, Orfanos P, Benetou V, Ntzani EE, Soerjomataram I, et al. Burden of hip fracture using disability‐adjusted life‐years: a pooled analysis of prospective cohorts in the CHANCES consortium. Lancet Public Health. 2017;2(5):e239–e246. [DOI] [PubMed] [Google Scholar]
- 3. Amarilla‐Donoso FJ, López‐Espuela F, Roncero‐Martín R, Leal‐Hernandez O, Puerto‐Parejo LM, Aliaga‐Vera I, et al. Quality of life in elderly people after a hip fracture: a prospective study. Health Qual Life Outcomes. 2020;18(1):71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Xu BY, Yan S, Low LL, Vasanwala FF, Low SG. Predictors of poor functional outcomes and mortality in patients with hip fracture: a systematic review. BMC Musculoskelet Disord. 2019;20(1):568. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Malafarina V, Reginster J‐Y, Cabrerizo S, Bruyère O, Kanis JA, Martinez JA, et al. Nutritional status and nutritional treatment are related to outcomes and mortality in older adults with hip fracture. Nutrients. 2018;10(5):555. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Bell JJ, Pulle RC, Crouch AM, Kuys SS, Ferrier RL, Whitehouse SL. Impact of malnutrition on 12‐month mortality following acute hip fracture. ANZ J Surg. 2016;86(3):157–161. [DOI] [PubMed] [Google Scholar]
- 7. Goisser S, Schrader E, Singler K, Bertsch T, Gefeller O, Biber R, et al. Malnutrition according to mini nutritional assessment is associated with severe functional impairment in geriatric patients before and up to 6 months after hip fracture. J Am Med Dir Assoc. 2015;16(8):661–667. [DOI] [PubMed] [Google Scholar]
- 8. Foo MXE, Wong GJY, Lew CCH. A systematic review of the malnutrition prevalence in hospitalized hip fracture patients and its associated outcomes. JPEN J Parenter Enteral Nutr. 2021;45(6):1141–1152. [DOI] [PubMed] [Google Scholar]
- 9. Detsky AS, McLaughlin JR, Baker JP, Johnston N, Whittaker S, Mendelson RA, et al. What is subjective global assessment of nutritional status? JPEN J Parenter Enteral Nutr. 1987;11(1):8–13. [DOI] [PubMed] [Google Scholar]
- 10. Stratton RJ, Hackston A, Longmore D, Dixon R, Price S, Stroud M, et al. Malnutrition in hospital outpatients and inpatients: prevalence, concurrent validity and ease of use of the ‘malnutrition universal screening tool’ (‘MUST’) for adults. Br J Nutr. 2004;92(5):799–808. [DOI] [PubMed] [Google Scholar]
- 11. Guigoz Y, Vellas B, Garry PJ. Assessing the nutritional status of the elderly: the mini nutritional assessment as part of the geriatric evaluation. Nutr Rev. 1996;54(1 Pt 2):S59–S65. [DOI] [PubMed] [Google Scholar]
- 12. Buzby GP, Mullen JL, Matthews DC, Hobbs CL, Rosato EF. Prognostic nutritional index in gastrointestinal surgery. Am J Surg. 1980;139(1):160–167. [DOI] [PubMed] [Google Scholar]
- 13. Onodera T, Goseki N, Kosaki G. Prognostic nutritional index in gastrointestinal surgery of malnourished cancer patients. Nihon Geka Gakkai Zasshi. 1984;85(9):1001–1005. [PubMed] [Google Scholar]
- 14. Tokunaga R, Sakamoto Y, Nakagawa S, Miyamoto Y, Yoshida N, Oki E, et al. Prognostic nutritional index predicts severe complications, recurrence, and poor prognosis in patients with colorectal cancer undergoing primary tumor resection. Dis Colon Rectum. 2015;58(11):1048–1057. [DOI] [PubMed] [Google Scholar]
- 15. Shoji F, Takeoka H, Kozuma Y, Toyokawa G, Yamazaki K, Ichiki M, et al. Pretreatment prognostic nutritional index as a novel biomarker in non‐small cell lung cancer patients treated with immune checkpoint inhibitors. Lung Cancer. 2019;136:45–51. [DOI] [PubMed] [Google Scholar]
- 16. Feng L, Chen W, Ping P, Ma T, Li Y, Xu L, et al. Preoperative malnutrition as an independent risk factor for the postoperative mortality in elderly Chinese individuals undergoing hip surgery: a single‐center observational study. Ther Adv Chronic Dis. 2022;13:20406223221102739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Ren H, Wu L, Hu W, Ye X, Yu B. Prognostic value of the c‐reactive protein/prognostic nutritional index ratio after hip fracture surgery in the elderly population. Oncotarget. 2017;8(37):61365–61372. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Zhang J, Yang M, Zhang X, He J, Wen L, Wang X, et al. The effectiveness of a co‐management care model on older hip fracture patients in China ‐ a multicentre non‐randomised controlled study. Lancet Reg Health West Pac. 2022;19:100348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Katzman R, Zhang MY, Ouang‐Ya‐Qu WZY, Wang Z, Liu WT, Yu E, et al. A Chinese version of the mini‐mental state examination; impact of illiteracy in a Shanghai dementia survey. J Clin Epidemiol. 1988;41(10):971–978. [DOI] [PubMed] [Google Scholar]
- 20. Kroenke K, Spitzer RL, Williams JB. The PHQ‐9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606–613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Herdman M, Gudex C, Lloyd A, Janssen M, Kind P, Parkin D, et al. Development and preliminary testing of the new five‐level version of EQ‐5D (EQ‐5D‐5L). Qual Life Res. 2011;20(10):1727–1736. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Luo N, Liu G, Li M, Guan H, Jin X, Rand‐Hendriksen K. Estimating an EQ‐5D‐5L value set for China. Value Health. 2017;20(4):662–669. [DOI] [PubMed] [Google Scholar]
- 23. Li S, Zhang J, Zheng H, Wang X, Liu Z, Sun T. Prognostic role of serum albumin, Total lymphocyte count, and mini nutritional assessment on outcomes after geriatric hip fracture surgery: a meta‐analysis and systematic review. J Arthroplasty. 2019;34(6):1287–1296. [DOI] [PubMed] [Google Scholar]
- 24. Cerri AP, Bellelli G, Mazzone A, Pittella F, Landi F, Zambon A, et al. Sarcopenia and malnutrition in acutely ill hospitalized elderly: prevalence and outcomes. Clin Nutr. 2015;34(4):745–751. [DOI] [PubMed] [Google Scholar]
- 25. Eglseer D, Halfens RJG, Lohrmann C. Is the presence of a validated malnutrition screening tool associated with better nutritional care in hospitalized patients? Nutrition. 2017;37:104–111. [DOI] [PubMed] [Google Scholar]
- 26. Zhang Y‐W, Cao M‐M, Li Y‐J, Dai G‐C, Lu P‐P, Zhang M, et al. Dietary protein intake in relation to the risk of osteoporosis in middle‐aged and older individuals: a cross‐sectional study. J Nutr Health Aging. 2022;26(3):252–258. [DOI] [PubMed] [Google Scholar]
- 27. Sim JH, Jun I‐G, Moon Y‐J, Jeon AR, Kim S‐H, Kim B, et al. Association of Preoperative Prognostic Nutritional Index and Postoperative Acute Kidney Injury in patients who underwent hepatectomy for hepatocellular carcinoma. J Pers Med. 2021;11(5):428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Ma S, Zhang B, Lu T, Li D, Li T, Shen Z, et al. Value of the prognostic nutritional index (PNI) in patients with newly diagnosed, CD5‐positive diffuse large B‐cell lymphoma: a multicenter retrospective study of the Huaihai lymphoma working group. Cancer. 2022;128(19):3487–3494. [DOI] [PubMed] [Google Scholar]
- 29. Wang D, Hu X, Xiao L, Long G, Yao L, Wang Z, et al. Prognostic nutritional index and systemic immune‐inflammation index predict the prognosis of patients with HCC. J Gastrointest Surg. 2021;25(2):421–427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Cheng Y‐L, Sung S‐H, Cheng H‐M, Hsu P‐F, Guo C‐Y, Yu W‐C, et al. Prognostic nutritional index and the risk of mortality in patients with acute heart failure. J Am Heart Assoc. 2017;6(6):e004876. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Feng L, Fu S, Yao Y, Yuan W, Zhao Y. Age, prognostic nutritional index, and Charlson comorbidity index were independent risk factors for postoperative Long‐term mortality in Chinese geriatric patients who sustain hip fracture. J Am Med Dir Assoc. 2021;22(12):2602–2603. [DOI] [PubMed] [Google Scholar]
- 32. Sze S, Zhang J, Pellicori P, Morgan D, Hoye A, Clark AL. Prognostic value of simple frailty and malnutrition screening tools in patients with acute heart failure due to left ventricular systolic dysfunction. Clin Res Cardiol. 2017;106(7):533–541. [DOI] [PubMed] [Google Scholar]
- 33. Wang Y, Jiang Y, Luo Y, Lin X, Song M, Li J, et al. Prognostic nutritional index with postoperative complications and 2‐year mortality in hip fracture patients: an observational cohort study. Int J Surg. 2023;109(11):3395–3406. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Munteanu C, Schwartz B. The relationship between nutrition and the immune system. Front Nutr. 2022;9:1082500. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The dataset is managed by Beijing Jishuitan Hospital. To request data access please contact the corresponding author.