Pre-Hospital Low ALT Activity, a Marker of Sarcopenia and Frailty, is Associated with Increased Long-Term Mortality after COPD Exacerbations. Historic Prospective Cohort Study.

Background COPD exacerbations have negative impact on patients' survival. Several risk factors for grave outcomes of such exacerbations have been descried. Muscle dysfunction and mass loss were shown to impact negatively on prognosis and survival. Low activity of the enzyme ALT (Alanine aminotransferase) in the blood is a known indicator for sarcopenia and frailty, however, no previous studies addressed the association of low ALT amongst patients hospitalized due to COPD exacerbation and long-term survival. Methods. This is a historic prospective cohort study of patients hospitalized due to acute COPD exacerbation. Results. Included were 232 consecutive COPD exacerbation patients. The median time of follow-up was 34.9 months (IQR 23.13 – 41.73 months). During this period 104 (44.8%) patients died. All patients were grouped to quartiles according to blood ALT levels (after exclusion of cases considered to have hepatitis (ALT > 40 IU)). The risk of long-term mortality increased, in a statistically signicant manner, amongst patients with low ALT values: the median survival of patients with ALT < 11 IU was 18.5 months only while the median survival for the rest of the study group was not reached. For ALT < 11IU; 12-16IU; 17-20IU and >21IU the mortality rates were 69%; 40.9%; 36.3% and 25% respectively (p < 0.001 for comparison of lower quartile with upper three quartiles). The crude hazard ratio for mortality amongst patients with ALT levels lower than 11IU was 2.37 (95% CI; 1.6 – 3.5). This increased risk of mortality remained signicant after adjustment for age, weight, creatinine, albumin concentration and cardiovascular diseases (HR = 1.83; 95% CI 1.08 – 3.1, p < 0.05). Conclusions. Low ALT values, a biomarker of sarcopenia and frailty, are associated with poor long-term survival amongst patients hospitalized due to COPD exacerbation.


Abstract
Background COPD exacerbations have negative impact on patients' survival. Several risk factors for grave outcomes of such exacerbations have been descried. Muscle dysfunction and mass loss were shown to impact negatively on prognosis and survival. Low activity of the enzyme ALT (Alanine aminotransferase) in the blood is a known indicator for sarcopenia and frailty, however, no previous studies addressed the association of low ALT amongst patients hospitalized due to COPD exacerbation and longterm survival. Methods. This is a historic prospective cohort study of patients hospitalized due to acute COPD exacerbation. Results. Included were 232 consecutive COPD exacerbation patients. The median time of follow-up was 34.9 months (IQR 23. 13 -41.73 months). During this period 104 (44.8%) patients died. All patients were grouped to quartiles according to blood ALT levels (after exclusion of cases considered to have hepatitis (ALT > 40 IU)). The risk of long-term mortality increased, in a statistically signi cant manner, amongst patients with low ALT values: the median survival of patients with ALT < 11 IU was 18.5 months only while the median survival for the rest of the study group was not reached. For ALT < 11IU; 12-16IU; 17-20IU and >21IU the mortality rates were 69%; 40.9%; 36.3% and 25% respectively (p < 0.001 for comparison of lower quartile with upper three quartiles). The crude hazard ratio for mortality amongst patients with ALT levels lower than 11IU was 2.37 (95% CI; 1.6 -3.5). This increased risk of mortality remained signi cant after adjustment for age, weight, creatinine, albumin concentration and cardiovascular diseases (HR = 1.83; 95% CI 1.08 -3.1, p < 0.05). Conclusions. Low ALT values, a biomarker of sarcopenia and frailty, are associated with poor long-term survival amongst patients hospitalized due to COPD exacerbation.

Background
Chronic obstructive pulmonary disease (COPD) is a highly prevalent disease worldwide. Acute COPD exacerbations have dismal results, both in high patients' mortality (ranging in different publications from over 5% to over 18% in-hospital mortality during exacerbation (1,2)) and in huge health expanses (1-6).
Several risk factors for grave outcomes of COPD exacerbations have been descried: Sakamoto and co.
(6) performed multivariable logistic regression analysis of over 3,000 COPD patients and found that older age and lower body mass index, amongst other patients' characteristics, are signi cantly associated with increased risk of in-hospital mortality. Serra-Picamal and co., in a big-data analysis of 17,555 COPD patients, identi ed early re-hospitalizations as another risk factor for mortality after hospitalization due to COPD exacerbation (7).
Survival and COPD mortality are negatively in uenced by both dysfunction and impaired muscle mass (sarcopenia) (8, 9) . Additionally, COPD exacerbations also rapidly induce loss of muscle mass and function through the activation of several biological pathways and systems. Sarcopenia and frailty, two overlapping syndromes, de ned as causing increased mortality and frequent hospital admissions, are prime candidates for having a negative impact on COPD exacerbations' outcomes. In a review published back in 2005, Morley and co. (11) stated that CODP is one of the chronic diseases particularly associated with frailty. Vellas and co. (12) relate to COPD as a causative agent for decreased muscle mass, in turn becoming a risk factor for COPD patients' mortality. In their review regarding "Sarcopenia and frailty in chronic respiratory disease", Bone and co. (13) declare the association between frailty and poor outcomes amongst COPD patients. Maddocks and co. (14) found that frailty affect 25% of all COPD patients and that it is a signi cant factor for pulmonary rehabilitation failure. Nevertheless, they state that pulmonary rehabilitation programs are bene cial for frail, COPD patients, potentially reversing frailty characteristics even with a short-term rehabilitation program.
Sarcopenia and the resultant frailty, represented by low activity of the enzyme ALT (SGPT) in the peripheral blood, were shown to be associated with shortened survival and other poor clinical outcomes, in various patients' populations and clinical scenarios. It is postulated that this association rise from the fact that low ALT levels are the result of sarcopenia and the resultant frailty is the cause for shortening of life span in low-ALT populations (15)(16)(17)(18).
However, there are no previous studies that examined the relationship between low ALT values and mortality amongst hospitalized patients with COPD exacerbation. The aim of the current study was to assess the association between ALT and clinical outcomes.

Patient selection
We executed a historic, prospective cohort study of COPD patients hospitalized to a single department of internal medicine, due to COPD exacerbation as their main diagnosis. The inclusion criteria were: a. age over 18 years, b. diagnosis of COPD exacerbation (ICD code 496) on discharge, c. available ALT blood levels, within the normal limits (lower than 40 IU/L) during the 3 months prior to admission. After study approval by an institutional ethics review board, we excluded patients with ALT levels greater than 40 IU/L, supposed to have hepatitis of any cause. We extracted patients' details from the computerized records, including epidemiologic characteristics, background diagnoses, long-term medications, laboratory parameters, short-and long-term clinical outcomes.

Statistical methods
In order to assess the association between ALT levels and dichotomous variables we calculated the sample size, as appropriate for medium strength associations. In order to determine such association we needed a sample size greater than 192 patients (for 1% signi cance and 80% power). The cohort size was determined using G Power software (19).
We assessed the distribution of continuous variables using histograms and Q-Q plots. We described continuous variables with normal distribution by means and standard deviations. Continuous variables that did not have normal distribution were described using median and IQR (Inter-quartile range). We described categorical variables using frequency and percentages. We assessed the association of clinical outcomes with continuous variables using t-tests or Man-Whitney tests. We assessed the association of categorical variables with clinical outcomes using the chi-square tests or the exact sher's test. We used logistic regression to assess the association between ALT values while correction of possible confounders. This was a two-block regression: the rst block included ALT values and the second block included possible confounders, selected by using the forward stepwise methods and the Wald test. Statistical analysis was done using the SPSS software (IBM Corp., NY USA). All statistical tests were twosided. We considered as signi cant only p values lower than 0.05.

Results
We included in this study 232 consecutive COPD patients, hospitalized due to COPD exacerbation in one internal-medicine department in a large, tertiary hospital, over a period of 38 months. After approval by an IRB, we extracted patients' data from their computerized medical records.  Table 2 describe patient characteristics found to be associated with increased risk for mortality amongst our study cohort. We divided all patients to quartiles according to their blood ALT levels (after exclusion of cases considered to have hepatitis (ALT > 40 IU)). The risk of long-term mortality increased, in a statistically signi cant manner, amongst patients with low ALT values: the median survival for patients with ALT < 11 IU was 18.5 months only while the median survival for the rest of the study group was not reached. For ALT < 11IU; 12-16IU; 17-20IU and >21IU the mortality rates were 69%; 40.9%; 36.3% and 25% respectively (p < 0.001 for comparison of lower quartile with upper three quartiles; gure 1).
The crude hazard ratio for mortality amongst patients with ALT levels lower than 11IU was 2.37 (95% CI; 1.6 -3.5). This increased risk of mortality remained signi cant after controlling for age, weight, Norton score, creatinine, albumin concentration and cardiovascular diseases (HR = 1.83; 95% CI 1.08 -3.1, p = 0.025). Table 3 describe the multivariate logistic regression for mortality in our study cohort.

Discussion
Current literature promotes personalized medicine. However, most cases of personalized medicine are actually dealing with precise medicine: diagnostic approaches become more focused on disease and pathology rather than on the person / patient as a whole. We believe that thorough evaluation of sarcopenia and frailty should be considered as true personalized medicine. Appreciation of sarcopenia and frailty enables a summation of patients' characteristics which are associated with his or her risk for subsequent morbidity and mortality. Addressing sarcopenia and frailty in quantitative measures should, therefore, better the prognostication process for many patients' populations. COPD patients should not be an outlier for the aforementioned declaration: these patients should be routinely evaluated for sarcopenia and frailty, and treatment strategies must take into account these patient's characteristics.
In the current study we show that similarly to other patients' groups, COPD patients can also be classi ed as sarcopenic and frail, according to their baseline ALT values. ALT measurements are a part of routine blood tests and are available worldwide for any hospitalized patient. Acknowledgment of low ALT values during COPD exacerbation could aid the attending physician in taking better clinical / therapeutic decisions (duration and route of administration of steroids, duration and type of antibiotics, promoting physiotherapy sessions and rehabilitation programs prior to patient's discharge to the community etc.).
ALT measurement should not replace other, well established methods for sarcopenia and frailty appreciation. Nevertheless, most such evaluation tools were developed and are being used in the realm of geriatric medicine. The results of this study are applicable to the general population of COPD patients admitted to internal-medicine departments. For these reasons, we suggest that ALT values would be agged, and diagnoses of sarcopenia and frailty should be highlighted in the discharge documents of COPD patients.

Conclusion
Low ALT values are an independent marker for shortened survival of COPD patients after being hospitalized due to COPD exacerbation. Being a marker for sarcopenia and frailty, low ALT values in these patients should prompt more aggressive therapeutic efforts on behalf of the attending physician. Future, prospective, interventional studies should aim at evaluation of different therapies and rehabilitation programs, concentrated at the sarcopenic and frail COPD patients.

Limitations
This was a retrospective analysis and therefor, not all confounding parameters could be addressed. We demonstrated statistical associations but causality could not be inferred. Also, this study was done in a single department and results might be changed with different patients' demographics.

Figure 1
Survival (months) of COPD patients according to baseline ALT values