Background/Purpose

There are few published studies on predictors of inpatient mortality in patients with systemic sclerosis (SSc). Knowledge of these predictors is important for early identification of patients at high-risk of inpatient death and recognition of potential modifiable factors. Currently available data are mostly from large databases which lack a granular view.  The aim of this study was to define factors associated with inpatient mortality in SSc.

Methods

All admissions coded for SSc (ICD-9 710.1) at the Hospital of University of Pennsylvania, between Jan 1, 2001 and Dec 31, 2011 were reviewed. The diagnosis of SSc was confirmed, and deaths were identified by chart review. For each death, an age (±5 years), sex and gender matched control with SSc (who did not die during their hospitalization) was identified. We hypothesized group differences in SSc characteristics, non–SSc co-morbidities and admission labs (as summarized in Table 1). Group differences were analyzed using Student’s t-test (and non parametric equivalents) as well as Chi²tests for dichotomous variables. Exposures associated with death in univariate analyses were used to form a final parsimonious multivariable logistic regression model.

Results

A total of 593 admissions and 30 deaths were identified. Data was not available on 1 death. A significant difference in non–SSc lung disease (p = 0.031), aspiration events (p < 0.001), WBC count (p = 0.048), Blood Urea Nitrogen (BUN) (p <0.001) and hemoglobin (Hb) (p = 0.025) was noted between subjects that died compared to matched controls. Odds of death were higher in patients with a higher BUN (OR = 1.06, CI = 1.02–1.11), non–SSc lung disease (OR = 3.87, CI = 1.26–11.88) and aspiration events (OR = 30, CI = 3.58–250.80), and lower in patients with a higher Hb (OR = 0.73, CI = 0.54–0.97). High BUN, a history of aspiration events and low Hb were found to be independently associated with mortality.

Conclusion

A high BUN on admission, low Hb, non–SSc Lung disease, and aspiration events are associated with higher in–hospital mortality in patients with SSc. The odds of dying in the hospital were 30 times in patients who aspirated. Therefore, stringent measures should be taken to prevent aspiration in these patients. Also, further study is needed to identify factors that predispose to acute aspiration events.

Table 1

Parameter	Death	Control	p – value
Transfer	6 of 29	8 of 19	0.76
Raynaud’s Phenomenon	22 of 23	22 of 23	1
Digital Ulcers	2 of 22	2 of 22	1
SSc – ILD	12 of 28	8 of 28	0.403
Pulmonary Artery HTN	16 of 22	14 of 22	0.747
SSc Reflux	22 of 25	22 of 25	1
Bacterial Overgrowth	2 of 19	3 of 19	1
SSc Renal Crisis	3 of 25	0 of 25	0.235
SSc – MSK	2 of 23	4 of 23	0.665
SSc – Cardiac	4 of 25	3 of 25	1
Coronary Artery Disease	19 of 27	13 of 27	0.176
Arrhythmia	13 of 27	9 of 27	0.277
Hypertension	23 of 29	18 of 29	0.370
Coronary Artery Disease	12 of 22	9 of 22	0.547
Chronic Renal Failure	9 of 29	4 of 29	0.207
Liver Disease	4 of 29	1 of 29	0.352
Diabetes Mellitus	7 of 29	2 of 29	0.144
Non SSc Lung Disease	16 of 29	7 of 29	0.031
Any Cause Lung Disease	18 of 29	14 of 29	0.429
Aspiration Events	15 of 29	1 of 29	< 0.001
Parameter	Mean (SD)	Mean (SD)	p – value
Hemoglobin	10.46 (1.72)	11.71 (2.31)	0.025
Hematocrit	31.89 (4.93)	35.43 (6.23)	0.022
WBC	12.98 (11.1)	8.5 (3.67)	0.048
BUN	37.23 (21.84)	19.77 (10.92)	<0.001
Creatinine	1.39 (0.71)	1.07 (0.78)	0.133
Sodium	137.86 (4.98)	138.25 (3.79)	0.741
Potassium	4.11 (0.69)	4.16 (0.66)	0.788
Platelets	256.36 (125.71)	228.75 (70.34)	0.315

Table 2

Univariate Logistic Regression
Parameter	Odds Ratio (95% CI)	p – value
Hemoglobin	0.73 (0.54 – 0.97)	0.032
BUN	1.06 (1.02 – 1.11)	0.004
Aspiration	30 (3.58 – 250.80)	0.002
Other Pulmonary	3.87 (1.26 – 11.88)	0.018
WBC	1.11 (0.99 – 1.25)	0.066
Multivariate Logistic Regression
Parameter	Odds Ratio (95% CI)	p – value
BUN	1.09 (1.02 – 1.17)	0.007
Hemoglobin	0.63 (0.43 – 0.93)	0.021
Aspiration	41.05 (3.62 – 464.91)	0.003

---

« Back to ACR/ARHP Annual Meeting 2014

ACR Meeting Abstracts - https://acrabstracts.org/abstract/predictors-of-inpatient-mortality-in-patients-with-systemic-sclerosis-a-case-control-study/