Abstract
Background
Idiopathic pleuroparenchymal fibroelastosis (IPPFE) is characterised by upper lobe-dominant fibrosis involving the pleura and subpleural lung parenchyma, with advanced cases often complicated by progressive weight loss. Therefore, we hypothesised that nutritional status is associated with mortality in IPPFE.
Methods
This retrospective study assesses nutritional status at the time of diagnosis and 1 year after diagnosis in 131 patients with IPPFE. Malnutrition-related risk was evaluated using the Geriatric Nutritional Risk Index (GNRI).
Results
Of the 131 patients, 96 (73.8%) were at malnutrition-related risk at the time of diagnosis according to the GNRI. Of these, 21 patients (16.0%) were classified as at major malnutrition-related risk (GNRI <82). Patients at major malnutrition-related risk were significantly older and had worse pulmonary function than patients at low (GNRI 92– <98) and moderate (GNRI 82– <92) malnutrition-related risk. GNRI scores decreased significantly from the time of diagnosis to 1 year after diagnosis. Patients with a lower GNRI (<91.8) had significantly shorter survival than patients with a median GNRI or higher (≥91.8). Patients with declines in annual GNRI scores of ≥5 had significantly shorter survival than patients with declines in annual GNRI scores of <5. In multivariate analysis, major malnutrition-related risk was significantly associated with increased mortality after adjustment for age, sex and forced vital capacity (hazard ratio 1.957). A composite scoring model including age, sex and major malnutrition-related risk was able to separate mortality risk in IPPFE.
Conclusion
Assessment of nutritional status by the GNRI provides useful information for managing patients with IPPFE by predicting mortality risk.
Short abstract
Malnutrition is frequent in patients with idiopathic pleuroparenchymal fibroelastosis and annual decreases in nutritional status are associated with increased mortality, indicating the importance of assessing nutritional status in this patient population https://bit.ly/3Z6PzMu
Introduction
Idiopathic pleuroparenchymal fibroelastosis (IPPFE) is a rare type of interstitial lung disease (ILD) characterised by fibrosis involving the pleura and subpleural lung parenchyma, predominantly in the upper lobes [1]. Patients with IPPFE typically present with dry cough and dyspnoea on exertion with decreased forced vital capacity (FVC) [2]. Importantly, the prognosis of patients with IPPFE has been reported to be equal to or worse than that of patients with idiopathic pulmonary fibrosis (IPF) [3, 4]. Although antifibrotic therapy has been used for patients with IPPFE, and may slow disease progression, its efficacy has not yet been determined [5–8]. Furthermore, there are currently no curative treatments for patients with IPPFE.
The assessment and improvement of nutritional status is important for improving outcomes in a range of diseases. Malnutrition is closely associated with progression of sarcopenia and cachexia, conditions commonly seen in patients with advanced lung disease. Importantly, malnutrition is not just a physical change, but also contributes to disease progression and worsens clinical outcomes. Indeed, the prognostic value of nutritional status has been validated in patients with a range of clinical conditions, including acute ischaemic stroke, heart failure, respiratory failure and malignancies. The majority of patients with IPPFE present with a lean body image and complain of weight loss associated with characteristic physical findings, including slender stature and a “flattened chest” [9]. However, few studies have evaluated nutritional status and its association with disease progression and outcomes in patients with IPPFE.
The Geriatric Nutritional Risk Index (GNRI) is a simple nutrition index calculated using serum albumin, body weight and ideal body weight [10]. The GNRI was originally developed to assess the risks of malnutrition and malnutrition-related mortality and morbidity in hospitalised patients. The utility of the GNRI has since been evaluated in a range of clinical conditions, including infectious and neoplastic diseases [11–14], and it is now used for nutritional assessment in a wide range of diseases [15–17]. The GNRI has been validated as a simple indicator of nutritional status that is more comprehensive than body mass index (BMI). However, there have been no reported studies evaluating nutritional status using the GNRI and its prognostic significance in patients with IPPFE. Therefore, the present study was conducted to evaluate nutritional status using the GNRI in patients with IPPFE and to investigate the association between GNRI scores and mortality.
Materials and methods
Patients
This retrospective study screened 146 consecutive patients with IPPFE who were admitted to Hamamatsu University Hospital, Seirei Hamamatsu Hospital and Seirei Mikatahara Hospital (Hamamatsu, Japan) between March 2004 and March 2021. 15 patients did not have height and weight data available. Thus, the present study enrolled a total of 131 patients with IPPFE. Patients were censored if they remained alive until 30 June 2022. The median (interquartile range (IQR)) observation period was 36.4 (20.9–69.3) months. The mortality rate was 58.8% during the observation period. The diagnosis of IPPFE was made according to the following criteria [18]: 1) PPFE radiographic pattern on chest computed tomography (CT) defined as bilateral subpleural dense consolidation with or without pleural thickening in the upper lobes and less marked or absent lower lobe involvement based on Reddy radiological criteria [19], with subpleural dense consolidation defined as consolidation below a line 1 cm from the apex of the lung (to exclude the pulmonary apical cap) with a minimum width of 1 cm in contact with pleura [18–22]; 2) radiological confirmation of disease progression, defined as an increase in upper lobe consolidation with or without pleural thickening and/or a decrease in upper lobe volume on serial radiological assessments; and 3) exclusion of other lung diseases with identifiable aetiologies, such as connective tissue disease-related ILD, chronic hypersensitivity pneumonitis, pulmonary sarcoidosis, pneumoconiosis and active pulmonary infection. The high-resolution CT patterns of lower lobe ILD were classified according to American Thoracic Society/European Respiratory Society/Japanese Respiratory Society/Latin American Thoracic Society IPF guidelines [23].
The study protocol was approved by the Ethical Committee of Hamamatsu University School of Medicine (22-108) and conducted in accordance with approved guidelines. The requirement for patient approval and/or informed consent was waived due to the retrospective study design.
Data collection
Clinical characteristics at the time of IPPFE diagnosis (age, sex, physical examination, smoking history, blood test results and pulmonary function test results) were retrieved from medical records.
GNRI assessments
GNRI scores were calculated based on data at the time of IPPFE diagnosis and 1 year after diagnosis as follows: GNRI=(1.489×serum albumin (g·L−1))+(41.7×(actual weight/ideal body weight)) [10]. Ideal body weight was calculated from the Lorentz equations (WLo) as: for men: (height (cm)−100)−((height (cm)−150)/4); for women: (height (cm)−100)−((height (cm)−150)/2.5) [10].
Originally, the GNRI was categorised into four levels: <82, major malnutrition-related risk; 82– <92, moderate malnutrition-related risk; 92– <98, low malnutrition-related risk; and ≥98, no malnutrition-related risk. GNRI stages in the present study were defined as at risk (<98 points) and not at risk (≥98 points) based on the total GNRI score.
Gender–age–physiology index
The gender–age–physiology (GAP) index was calculated on the basis of data at the time of IPPFE diagnosis, as previously described [24]: sex (female, 0 points; male, 1 point), age (≤60 years, 0 points; 61–65 years, 1 point; >65 years, 2 points), FVC (% pred) (>75%, 0 points; 50–75%, 1 point; <50%, 2 points) and diffusing capacity of the lung for carbon monoxide (DLCO) (% pred) (>55%, 0 points; 36–55%, 1 point; ≤35%, 2 points; cannot perform, 3 points). The GAP index was defined based on the total GAP score: stage I (0–3 points), stage II (4–5 points) and stage III (6–8 points).
Composite model comprising gender, age and malnutrition-related risk
A composite model was generated to assess mortality risk based on the GAP model with gender, age and physiology as variables. The presence of major malnutrition-related risk was used to replace the physiology variable in the GAP index. Thus, a composite model was created using gender, age and presence of major malnutrition-related risk, as age, male sex and presence of major malnutrition-related risk were independently associated with increased mortality. Age >60 years and male sex were both included as a risk factor based on the GAP index. A leave-one-out analysis was performed to avoid overfitting of the variables (supplementary figure S1). A simple scoring system was developed; 1 point was assigned if the patient's age was >60 years, the patient was male sex or had major malnutrition-related risk (GNRI <82). Accordingly, patients were categorised into three groups based on the total point scores: mild (0–1), moderate (2) and severe (3). The model was evaluated using Harrell's concordance index (C-index).
Statistical analysis
Discrete variables are presented as count (percentage) and continuous variables are presented as median (IQR). The Mann–Whitney U-test and the Wilcoxon matched signed-rank test were used to compare unmatched and matched continuous variables, respectively. Fisher's exact test for independence was used to compare categorical variables. Cox proportional hazard regression analysis was used to identify factors associated with mortality. Among the statistically significant covariates identified in univariate analysis, clinically relevant and important variables (age, sex and FVC (% pred)) were selected for inclusion in the multivariate analysis. Cumulative survival probabilities were estimated using the Kaplan–Meier method and log-rank test. Overall survival time was measured from the date of IPPFE diagnosis or GNRI assessment at 1 day after diagnosis. Cut-offs for the GNRI score at diagnosis were determined according to the median GNRI score or the presence of major malnutrition-related risk, respectively. Cut-offs for annual changes in the GNRI were temporally determined accordingly to the first tertile and clinically meaningful time-points. All statistical analyses were conducted using R version 4.11 [25]. All hypothesis tests were two-tailed. p-values <0.05 were considered statistically significant.
Results
Clinical characteristics
Patient clinical characteristic are shown in table 1. The median age was 66 years and 86 patients (65.6%) were male. Approximately 64% of patients were never-smokers. Physical examination demonstrated a median BMI of 17.2 kg·m−2 and a median FVC of 64.7% predicted, indicating moderate-to-severe reductions in FVC. Lower lobe ILD on chest CT was observed in 81 patients (61.8%). The median serum total protein and albumin levels were 7.3 g·L−1 and 4.0 g·L−1, respectively.
TABLE 1.
Clinical characteristics of patients with idiopathic pleuroparenchymal fibroelastosis (n=131)
| Age, years | 66 (63–76) |
| Sex | |
| Male | 86 (65.6) |
| Female | 45 (34.4) |
| Observation period, months | 36.4 (20.9–69.3) |
| Mortality | 77 (58.8) |
| Smoking status | |
| Never-smoker | 84 (64.1) |
| Ex-smoker | 47 (35.9) |
| Pack-years | 0 (0–10) |
| BMI, kg·m−2 | 17.2 (14.8–18.6) |
| Flat chest | 33 (25.2) |
| Pulmonary function tests | |
| FVC, % pred | 64.7 (48.2–80.4) |
| FEV1, % pred | 77.9 (60.9–95.5) |
| FEV1/FVC, % | 95.7 (89.4–100) |
| DLCO, % | 91.2 (71.3–112.9) (n=84) |
| RV/TLC, % | 48.3 (42.5–56.9) (n=82) |
| CT imaging | |
| Presence of lower lobe ILD | 81 (61.8) |
| UIP pattern | |
| Definite | 14 |
| Probable | 35 |
| Indeterminate | 25 |
| Alternative | 7 |
| Laboratory tests | |
| PaO2, Torr | 80.2 (71.4–87.9) (n=98) |
| PaCO2, Torr | 46.5 (41.3–50.9) (n=98) |
| KL-6, U·mL−1 | 436 (331–609) (n=128) |
| SP-D, ng·mL−1 | 172 (118–252) (n=126) |
| LDH, U·L−1 | 193 (173–215) |
| TP, g·dL−1 | 7.3 (7.1–7.7) |
| Alb, g·dL−1 | 4.0 (3.7–4.3) |
| Antifibrotic therapy | |
| Pirfenidone | 12 |
| Nintedanib | 0 |
Data are presented as median (interquartile range), n (%) or n. BMI: body mass index; FVC: forced vital capacity; FEV1: forced expiratory volume in 1 s; DLCO: diffusing capacity of the lung for carbon monoxide; RV: residual volume; TLC: total lung capacity; CT: computed tomography; ILD: interstitial lung disease; UIP: usual interstitial pneumonia; PaO2: arterial oxygen tension; PaCO2: arterial carbon dioxide tension; KL-6: Krebs von den Lungen-6; SP-D: surfactant protein D; LDH: lactate dehydrogenase; TP: total protein; Alb: albumin.
Assessment of malnutrition-related risk according to the GNRI
The GNRI score distribution in patients with IPPFE is shown in figure 1. The median (IQR) GNRI score was 91.8 (84.1–97.4). Only 31 patients (23.7%) were classified as “no malnutrition-related risk (GNRI ≥98)”, with 24.4%, 35.9% and 16.0% of patients classified as “low malnutrition-related risk (GNRI 92– <98)”, “moderate malnutrition-related risk (GNRI 82– <92)” and “major malnutrition-related risk (GNRI <82)”, respectively.
FIGURE 1.
Distribution of Geriatric Nutritional Risk Index (GNRI) scores and malnutrition-related risk in patients with idiopathic pleuroparenchymal fibroelastosis at the time of diagnosis.
Patients at major malnutrition-related risk were significantly older and had poorer pulmonary function than patients at low or moderate malnutrition-related risk (supplementary table S1).
Association between nutritional status and survival
We next assessed survival according to malnutrition-related risk defined by the GNRI. Patients with a lower median GNRI (<91.8) had significantly shorter survival than patients with a median GNRI or higher (≥91.8) (figure 2a). No significant difference in survival was observed between patients at no malnutrition-related risk and those at low malnutrition-related risk. However, patients at low malnutrition-related risk had significantly longer survival than patients at moderate and major malnutrition-related risk (figure 2b). Patients at major malnutrition-related risk had significantly shorter survival compared with the other groups (figure 2c).
FIGURE 2.
Association of malnutrition-related risk and mortality in patients with idiopathic pleuroparenchymal fibroelastosis (IPPFE). a) Kaplan–Meier curves of patients with IPPFE according to Geriatric Nutritional Risk Index (GNRI) scores above and below the median. b) Kaplan–Meier curves of patients with malnutrition-related risk stratified according to GNRI score. c) Kaplan–Meier curves of patients with presence or absence of major malnutrition-related risk determined by GNRI score. p-values were determined by the log-rank test.
Univariate and multivariate analysis of the GNRI for mortality
We determined whether malnutrition-related risk defined by the GNRI and GNRI score were associated with mortality. Univariate analysis demonstrated that age, sex, FVC, serum albumin, presence of lower lobe ILD, lower GNRI score and presence of major malnutrition-related risk were significantly associated with increased mortality. BMI was not associated with mortality (table 2). Multivariate analysis revealed that presence of major malnutrition-related risk was significantly associated with increased mortality independent of age, sex and FVC (table 2). GNRI score did not remain significant in multivariate analysis.
TABLE 2.
Univariate and multivariate Cox proportion analysis for mortality in patients with idiopathic pleuroparenchymal fibroelastosis
| Predictor | Hazard ratio (95% CI) | p-value |
| Univariate analysis | ||
| Age, years | 1.060 (1.034–1.087) | <0.001 |
| Male | 2.251 (1.357–3.906) | 0.003 |
| BMI, kg·m−2 | 0.964 (0.886–1.048) | 0.395 |
| Flat chest | 0.933 (0.576–1.564) | 0.784 |
| FVC, % pred | 0.973 (0.962–0.984) | <0.001 |
| FEV1, % pred | 0.981 (0.972–0.991) | <0.001 |
| FEV1/FVC, % | 1.093 (1.045–1.148) | <0.001 |
| DLCO, % | 0.983 (0.973–0.992) | <0.001 |
| RV/TLC, % | 1.069 (1.036–1.104) | <0.001 |
| KL-6, U·mL−1 | 1.001 (1.000–1.001) | <0.001 |
| SP-D, ng·mL−1 | 1.001 (1.000–1.002) | 0.005 |
| TP, g·dL−1 | 0.866 (0.590–1.270) | 0.462 |
| Alb, g·dL−1 | 0.353 (0.217–0.582) | <0.001 |
| LDH, U·L−1 | 1.009 (1.004–1.015) | <0.001 |
| Lower lobe ILD | 3.169 (1.879–5.641) | <0.001 |
| GNRI, continuous variables | 0.951 (0.927–0.976) | <0.001 |
| GNRI, per malnutrition-related risk | 1.518 (1.185–1.960) | 0.001 |
| Major malnutrition-related risk: GNRI <82 | 2.892 (1.544–5.129) | <0.001 |
| Multivariate analysis 1 | ||
| Age, years | 1.049 (1.019–1.080) | 0.001 |
| Male | 4.496 (2.502–8.532) | <0.001 |
| FVC, % pred | 0.970 (0.958–0.982) | <0.001 |
| Major malnutrition-related risk: GNRI <82 | 1.957 (1.024–4.000) | 0.039 |
| Multivariate analysis 2 | ||
| Age, years | 1.047 (1.018–1.079) | 0.002 |
| Male | 4.414 (2.469–8.334) | <0.001 |
| FVC, % pred | 0.971 (0.958–0.983) | <0.001 |
| GNRI, continuous variables | 0.979 (0.950–1.011) | 0.183 |
BMI: body mass index; FVC: forced vital capacity; FEV1: forced expiratory volume in 1 s; DLCO: diffusing capacity of the lung for carbon monoxide; RV: residual volume; TLC: total lung capacity; KL-6: Krebs von den Lungen-6; SP-D: surfactant protein D; TP: total protein; Alb: albumin; LDH: lactate dehydrogenase; ILD: interstitial lung disease; GNRI: Geriatric Nutritional Risk Index.
DLCO was evaluated in 84 patients with IPPFE. When the GNRI was evaluated together with age, sex, FVC and DLCO by multivariate analyses, the values of the GNRI and DLCO were not significant (supplementary table S2).
Longitudinal assessment of malnutrition-related risks in patients with IPPFE
Next, we examined the association between mortality and annual changes in nutritional status as assessed by the GNRI. Among 131 patients, 109 patients had nutritional assessments both at the time of diagnosis and 1 year later. GNRI scores decreased significantly from the time of diagnosis to 1 year after diagnosis (91.2 versus 89.7, respectively; p=0.002) (figure 3a), with 32 patients (29.4%) found to have a marked decrease in GNRI score (decrease of ≥5). Patients whose GNRI scores declined by ≥5 had significantly short survival than patients whose GNRI scores declined by <5 (figure 3b).
FIGURE 3.
Annual changes in nutritional status and association with mortality in patients with idiopathic pleuroparenchymal fibroelastosis (IPPFE). a) Annual changes in Geriatric Nutritional Risk Index (GNRI) score in patients with IPPFE at the time of diagnosis and at 1 year after diagnosis. Median (interquartile range) values are indicated. b) Kaplan–Meier curves of patients with IPPFE according to annual changes in GNRI score. p-values were determined by the log-rank test.
Mortality risk according to age, sex and presence of major malnutrition-related risk in patients with IPPFE
The GAP model has been widely used and validated as a mortality risk assessment for IPF patients. Therefore, we first attempted to apply the GAP model to assess mortality risk in IPPFE patients. As shown in figure 4a, the GAP model showed poor performance in discriminating mortality. In particular, stage II (moderate) and stage III (severe) survival curves were reversed, with patients with stage II found to have worse survival than patients with stage III.
FIGURE 4.
Kaplan–Meier curves of patients with idiopathic pleuroparenchymal fibroelastosis (IPPFE) based on age, sex and major malnutrition-related risk. a) Kaplan–Meier curves of patients with IPPFE according to the gender–age–physiology index. b) Kaplan–Meier curves of patients with IPPFE and age, sex and presence of major malnutrition-related risk determined by the Geriatric Nutritional Risk Index. p-values were determined by the log-rank test.
We next attempted to develop a composite model for assessing mortality risk using major malnutrition-related risk together with age and sex, as age, sex and presence of major malnutrition-related risk were independently associated with increased mortality. In line with the GAP index, 1 point was assigned for age >60 years, male sex and presence of major malnutrition-related risk. Patients were categorised into three groups based on total point scores: mild (0–1), moderate (2) and severe (3). Our composite model demonstrated good prognostic separation (median survival times: mild, 24.8 months; moderate, 38.8 months; and severe, 98.5 months; C-index 0.719) (figure 4b).
Discussion
The present study assessed nutritional status using the GNRI and evaluated its clinical significance in patients with IPPFE. We found that >75% of patients with IPPFE were considered to be at malnutrition-related risk (GNRI <98) at the time of diagnosis, with major malnutrition-related risk (GNRI <82) identified in 16% of patients. Nutritional status assessed by the GNRI significantly decreased 1 year after diagnosis. Importantly, patients with lower GNRI scores had significantly shorter survival compared with patients with higher GNRI scores. In addition, major malnutrition-related risk was significantly associated with increased mortality in multivariate analysis independent of age, sex and FVC. Moreover, annual decline in nutritional status assessed by the GNRI was significantly associated with shorter survival and higher mortality. A simple composite model with age, sex and major malnutrition-related risk yielded good prognostic separation in IPPFE patients. Collectively, these results suggest that the majority of patients with IPPFE have poor nutritional status based on GNRI scores and that assessment of nutritional status by the GNRI has utility in predicting outcomes in IPPFE patients.
The GNRI, consisting of BMI and serum albumin levels, is a valid tool for assessing malnutrition-related morbidity [10] and mortality in patients with various clinical conditions, including acute ischaemic stroke, heart failure, respiratory failure and malignancies [13–17]. However, there have been no studies using the GNRI to assess the nutritional status of patients with IPPFE who often have lower BMI and slender body types. The present study is the first to evaluate the GNRI in patients with IPPFE. Several comprehensive nutrition scoring systems have been developed to assess nutritional status. However, these scoring systems are typically complex, requiring multiple items to be calculated. In contrast, the GNRI is a simple nutrition scoring system requiring only BMI and serum albumin levels, both of which are routinely and easily measured in clinical practice. The present study clearly demonstrated that three-quarters of patients with IPPFE had malnutrition-related risk according to the GNRI, with nutritional status significantly deteriorating at 1 year after diagnosis. Taken together, these observations indicate that the majority of patients with IPPFE are malnourished at the time of diagnosis and that their nutritional status worsens over time. Indeed, nutritional status, indeterminate efficacy and a high prevalence of gastrointestinal disorders due to antifibrotic therapy might have had an effect on the small number of patients treated with antifibrotic therapy in this study.
Importantly, the present study demonstrated that poor nutritional status defined by the GNRI was associated with mortality risk in patients with IPPFE. Indeed, patients with a lower median GNRI (<91.8) had significantly poorer survival than patients with a median GNRI or higher (≥91.8). In addition, multivariate Cox regression hazard analysis identified major malnutrition-related risk as a significant prognostic factor independent of age, sex and FVC. However, neither major malnutrition-related risk nor GNRI score were significant when evaluated with age, FVC and DLCO. This can be attributed to the limited number of patients evaluated for DLCO in the present study. Although BMI is also thought to partially represent nutritional status, BMI was not a significant prognostic factor even in univariate Cox regression hazard analysis in our cohort of patients with IPPFE. Our results are consistent with previous studies that examined the association between BMI and mortality risk by Cox regression hazard analyses in patients with IPPFE [26–28]. The reason for this discrepancy in the prognostic value between the GNRI and BMI may be that the GNRI is more comprehensive than BMI in assessing nutritional status in IPPFE, and the prevalence of BMI in this study was distributed in a narrow range (IQR 14.8–18.6 kg·m−2) and as low as 17.2 kg·m−2.
In addition, we found that longitudinal change in nutritional status assessed by the GNRI was associated with mortality risk in IPPFE patients. Patients who had a greater decline in the GNRI had a significantly poorer prognosis than patients who did not, suggesting that trends in the GNRI are also important in predicting the prognosis of patients with IPPFE. Collectively, these observations suggest that the GNRI at the time of diagnosis and the trend in the GNRI over time are significant prognostic factors in patients with IPPFE. Malnutrition in IPPFE patients represents a multifactorial problem reflecting the complex interplay of systematic inflammation leading to cachexia, worsening nutritional status from poor intake, and reduced exercise tolerance due to sarcopenia and dyspnoea. Indeed, our previous study reported that muscle wasting is frequently observed and body composition change evaluated by elector spine muscle attenuation on CT is associated with mortality in patients with IPPFE [26]. In line with our results, interestingly, body weight loss was strongly correlated with FVC decline in patients with IPPFE and IPF [29, 30]. These results suggest an underlying mechanism between the pathogenesis of lung fibrosis and deterioration of nutritional status. Additionally, short-term efficacy of pulmonary rehabilitation in terms of exercise capacity was also reported in patients with IPPFE [31]. Therefore, it is of great interest to determine whether direct interventions to improve nutritional status using a nutrition support team, supplements or anamorelin (a ghrelin receptor agonist) can improve clinical outcomes in patients with IPPFE.
To date, several prognostic factors including lower FVC [27, 28, 32, 33], presence of lower lobe ILD [28] and history of pneumothorax [34] have been reported in patients with IPPFE [34]. Recently, we also identified upper lobe lung volume measured using three-dimensional (3D)-CT and standardised with predicted FVC as a significant prognostic factor in patients with IPPFE [32]. However, the significance of these reported prognostic factors is not fully consistent between previous studies. When predicting the prognosis of patients with a complex disease such as IPPFE, the use of a single modality may be insufficient. Instead, a composite model that includes multiple measurements would be preferable. In this regard, we attempted to develop a composite model using age, gender and presence of major malnutrition-related risk, all of which were found to independent poor prognostic factors in the present study. We found that our composite model had good prognostic discrimination in patients with IPPFE (C-index 0.719). In contrast, the GAP model performed relatively poorly in discriminating IPPFE prognosis, which is consistent with the results of our previous study. Indeed, survival curves for GAP stage II and GAP stage III were completely inverted in the present study. Recently, we reported a separate composite model using age, sex and standardised upper lobe lung volumes measured by 3D-CT [32]. This 3D-CT model achieved good prognostic separation with a higher C-index (0.762) in patients with IPPFE. However, the use of a 3D-CT composite model is time consuming and costly, and has the additional weakness of radiation exposure. In contrast, the present composite model is simple and less time consuming with no radiation exposure. Taken together, our simple composite model using age, sex and presence of major malnutrition-related risk defined by the GNRI represents a useful tool for assessing mortality risk in patients with IPPFE during routine clinical practice.
The present study had several limitations. First, this was a retrospective cohort study and the number of patients was relatively small as IPPFE is a rare type of ILD. Additionally, all of the patients were Asian. In particular, the BMI of Asian patients tends to be lower than those of other ethnicities. Therefore, prospective international validation studies involving a larger number of patients are required to confirm the results of the present study. Second, although this study assessed annual changes in nutritional status among patients with IPPFE, the observation period was relatively short. Therefore, future longer term studies are required to fully assess changes in nutritional status over time among patients with IPPFE. Third, the present study assessed nutrition status using the GNRI only; however, there are several methods of evaluating nutritional status.
In conclusion, the results of the present study demonstrate that patients with IPPFE frequently have poor and progressively worsening malnutrition as assessed by the GNRI. Importantly, nutritional status and annual change in the GNRI were significantly associated with increased risk of mortality. Our composite scoring model, including age, sex and presence of major malnutrition-related risk as defined by the GNRI, achieved good prognostic separation in patients with IPPFE, indicating that this model has utility in predicting mortality. Collectively, these results indicate that assessment of nutritional status by the GNRI provides useful information for managing IPPFE by estimating mortality risk in clinical practice.
Acknowledgements
We thank Masahiro Shirai, Kazuhiro Asada and Keigo Koda for data collection. We thank Enago (www.enago.com) for editing a draft of the manuscript.
Provenance: Submitted article, peer reviewed.
Data availability: The data that support the findings of this study are available from the corresponding author upon reasonable request.
Author contributions: Y. Suzuki: conception and design, data collection, data analysis and interpretation, manuscript writing, and final approval of the manuscript. A. Fukada: data collection. K. Mori: statistical analysis. M. Kono, H. Hasegawa, D. Hashimoto, K. Yokomura and S. Imokawa: conception and design, data collection, and data analysis. Y. Inoue, H. Yasui, H. Hozumi, M. Karayama, K. Furuhashi, N. Enomoto, T. Fujisawa, N. Inui and H. Nakamura: data collection, data analysis and supervision. T. Suda: conception and design, manuscript writing, and administrative support.
Conflicts of interest: The authors declare that no competing interests exist.
Support statement: This work was supported by a grant-in-aid for scientific research from the Japan Society for the Promotion of Science (grant number 22K08279 received by Y. Suzuki).
References
- 1.Travis WD, Costabel U, Hansell DM, et al. An official American Thoracic Society/European Respiratory Society statement: update of the international multidisciplinary classification of the idiopathic interstitial pneumonias. Am J Respir Crit Care Med 2013; 188: 733–748. doi: 10.1164/rccm.201308-1483ST [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Chua F, Desai SR, Nicholson AG, et al. Pleuroparenchymal fibroelastosis. A review of clinical, radiological, and pathological characteristics. Ann Am Thorac Soc 2019; 16: 1351–1359. doi: 10.1513/AnnalsATS.201902-181CME [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Fujisawa T, Mori K, Mikamo M, et al. Nationwide cloud-based integrated database of idiopathic interstitial pneumonias for multidisciplinary discussion. Eur Respir J 2019; 53: 1802243. doi: 10.1183/13993003.02243-2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Suzuki Y, Fujisawa T, Sumikawa H, et al. Disease course and prognosis of pleuroparenchymal fibroelastosis compared with idiopathic pulmonary fibrosis. Respir Med 2020; 171: 106078 doi: 10.1016/j.rmed.2020.106078 [DOI] [PubMed] [Google Scholar]
- 5.Nasser M, Si-Mohamed S, Turquier S, et al. Nintedanib in idiopathic and secondary pleuroparenchymal fibroelastosis. Orphanet J Rare Dis 2021; 16: 419. doi: 10.1186/s13023-021-02043-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sugino K, Ono H, Shimizu H, et al. Treatment with antifibrotic agents in idiopathic pleuroparenchymal fibroelastosis with usual interstitial pneumonia. ERJ Open Res 2021; 7: 00196-2020. doi: 10.1183/23120541.00196-2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kinoshita Y, Miyamura T, Ikeda T, et al. Limited efficacy of nintedanib for idiopathic pleuroparenchymal fibroelastosis. Respir Investig 2022; 60: 562–569. doi: 10.1016/j.resinv.2022.03.001 [DOI] [PubMed] [Google Scholar]
- 8.Cottin V, Si-Mohamed S, Diesler R, et al. Pleuroparenchymal fibroelastosis. Curr Opin Pulm Med 2022; 28: 432–440. doi: 10.1097/MCP.0000000000000907 [DOI] [PubMed] [Google Scholar]
- 9.Wataneba K. Pleuroparenchymal fibroelastosis: its clinical characteristics. Curr Respir Med Rev 2013; 9: 229–237. doi: 10.2174/1573398X0904140129125307 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bouillanne O, Morineau G, Dupont C, et al. Geriatric Nutritional Risk Index: a new index for evaluating at-risk elderly medical patients. Am J Clin Nutr 2005; 82: 777–783. doi: 10.1093/ajcn/82.4.777 [DOI] [PubMed] [Google Scholar]
- 11.Lee JS, Choi HS, Ko YG, et al. Performance of the Geriatric Nutritional Risk Index in predicting 28-day hospital mortality in older adult patients with sepsis. Clin Nutr 2013; 32: 843–848. doi: 10.1016/j.clnu.2013.01.007 [DOI] [PubMed] [Google Scholar]
- 12.Wei L, Xie H, Li J, et al. The prognostic value of geriatric nutritional risk index in elderly patients with severe community-acquired pneumonia: a retrospective study. Medicine 2020; 99: e22217. doi: 10.1097/MD.0000000000022217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Karayama M, Inoue Y, Yasui H, et al. Association of the Geriatric Nutritional Risk Index with the survival of patients with non-small-cell lung cancer after platinum-based chemotherapy. BMC Pulm Med 2021; 21: 409. doi: 10.1186/s12890-021-01782-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Karayama M, Inoue Y, Yoshimura K, et al. Association of the Geriatric Nutritional Risk Index with the survival of patients with non-small cell lung cancer after nivolumab therapy. J Immunother 2022; 45: 125–131. doi: 10.1097/CJI.0000000000000396 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kang MK, Kim TJ, Kim Y, et al. Geriatric nutritional risk index predicts poor outcomes in patients with acute ischemic stroke – automated undernutrition screen tool. PLoS One 2020; 15: e0228738. doi: 10.1371/journal.pone.0228738 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sze S, Pellicori P, Kazmi S, et al. Prevalence and prognostic significance of malnutrition using 3 scoring systems among outpatients with heart failure: a comparison with body mass index. JACC Heart Fail 2018; 6: 476–486. doi: 10.1016/j.jchf.2018.02.018 [DOI] [PubMed] [Google Scholar]
- 17.Yenibertiz D, Cirik MO. The comparison of GNRI and other nutritional indexes on short-term survival in geriatric patients treated for respiratory failure. Aging Clin Exp Res 2021; 33: 611–617. doi: 10.1007/s40520-020-01740-8 [DOI] [PubMed] [Google Scholar]
- 18.Enomoto Y, Nakamura Y, Satake Y, et al. Clinical diagnosis of idiopathic pleuroparenchymal fibroelastosis: a retrospective multicenter study. Respir Med 2017; 133: 1–5. doi: 10.1016/j.rmed.2017.11.003 [DOI] [PubMed] [Google Scholar]
- 19.Reddy TL, Tominaga M, Hansell DM, et al. Pleuroparenchymal fibroelastosis: a spectrum of histopathological and imaging phenotypes. Eur Respir J 2012; 40: 377–385. doi: 10.1183/09031936.00165111 [DOI] [PubMed] [Google Scholar]
- 20.Bonifazi M, Montero MA, Renzoni EA. Idiopathic pleuroparenchymal fibroelastosis. Curr Pulmonol Rep 2017; 6: 9–15. doi: 10.1007/s13665-017-0160-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sumikawa H, Johkoh T, Egashira R, et al. Pleuroparenchymal fibroelastosis-like lesions in patients with interstitial pneumonia diagnosed by multidisciplinary discussion with surgical lung biopsy. Eur J Radiol Open 2020; 7: 100298. doi: 10.1016/j.ejro.2020.100298 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Fujisawa T, Horiike Y, Egashira R, et al. Radiological pleuroparenchymal fibroelastosis-like lesion in idiopathic interstitial pneumonias. Respir Res 2021; 22: 290. doi: 10.1186/s12931-021-01892-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Raghu G, Remy-Jardin M, Myers JL, et al. Diagnosis of idiopathic pulmonary fibrosis. An official ATS/ERS/JRS/ALAT clinical practice guideline. Am J Respir Crit Care Med 2018; 198: e44–e68. doi: 10.1164/rccm.201807-1255ST [DOI] [PubMed] [Google Scholar]
- 24.Ley B, Ryerson CJ, Vittinghoff E, et al. A multidimensional index and staging system for idiopathic pulmonary fibrosis. Ann Intern Med 2012; 156: 684–691. doi: 10.7326/0003-4819-156-10-201205150-00004 [DOI] [PubMed] [Google Scholar]
- 25.Kanda Y. Investigation of the freely available easy-to-use software ‘EZR’ for medical statistics. Bone Marrow Transplant 2013; 48: 452–458. doi: 10.1038/bmt.2012.244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Suzuki Y, Yoshimura K, Enomoto Y, et al. Distinct profile and prognostic impact of body composition changes in idiopathic pulmonary fibrosis and idiopathic pleuroparenchymal fibroelastosis. Sci Rep 2018; 8: 14074. doi: 10.1038/s41598-018-32478-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Shioya M, Otsuka M, Yamada G, et al. Poorer prognosis of idiopathic pleuroparenchymal fibroelastosis compared with idiopathic pulmonary fibrosis in advanced stage. Can Respir J 2018; 2018: 6043053. doi: 10.1155/2018/6043053 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kono M, Fujita Y, Takeda K, et al. Clinical significance of lower-lobe interstitial lung disease on high-resolution computed tomography in patients with idiopathic pleuroparenchymal fibroelastosis. Respir Med 2019; 154: 122–126. doi: 10.1016/j.rmed.2019.06.018 [DOI] [PubMed] [Google Scholar]
- 29.Kono M, Tsunoda T, Ikeda S, et al. Clinical features of idiopathic pleuroparenchymal fibroelastosis with progressive phenotype showing a decline in forced vital capacity. Respir Investig 2023; 61: 210–219. doi: 10.1016/j.resinv.2023.01.003 [DOI] [PubMed] [Google Scholar]
- 30.Jouneau S, Crestani B, Thibault R, et al. Analysis of body mass index, weight loss and progression of idiopathic pulmonary fibrosis. Respir Res 2020; 21: 312. doi: 10.1186/s12931-020-01528-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Mori Y, Yamano Y, Kataoka K, et al. Pulmonary rehabilitation for idiopathic pleuroparenchymal fibroelastosis: a retrospective study on its efficacy, feasibility, and safety. Respir Investig 2021; 59: 849–858. doi: 10.1016/j.resinv.2021.08.003 [DOI] [PubMed] [Google Scholar]
- 32.Fukada A, Suzuki Y, Mori K, et al. Idiopathic pleuroparenchymal fibroelastosis: three-dimensional computed tomography assessment of upper-lobe lung volume. Eur Respir J 2022; 60: 2200637. doi: 10.1183/13993003.00637-2022 [DOI] [PubMed] [Google Scholar]
- 33.Kinoshita Y, Ikeda T, Miyamura T, et al. A proposed prognostic prediction score for pleuroparenchymal fibroelastosis. Respir Res 2021; 22: 215. doi: 10.1186/s12931-021-01810-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kono M, Nakamura Y, Enomoto Y, et al. Pneumothorax in patients with idiopathic pleuroparenchymal fibroelastosis: incidence, clinical features, and risk factors. Respiration 2021; 100: 19–26. doi: 10.1159/000511965 [DOI] [PubMed] [Google Scholar]