Introduction: Hematopoietic stem cell transplantation (HSCT) is an effective treatment option for patients with hematological and neoplastic diseases. Despite recent improvements, HSCT is still associated with a significant risk of mortality. Determining risk factors for death within the first 100 days after HSCT could help to identify patients who would benefit from interventions in order to decrease that risk. We have previously reported that increased fluid accumulation during conditioning chemotherapy and in the early period after HSCT was associated with decreased survival (Costa EMM et al, ASH 2013Abstract #4512). Herein, we expand that series of patients and we further analyze the impact of weight gain on early (100-days) mortality in patients who underwent HSCT at our institution.

Objective: To determine the impact of weight gain during the first 10 days post-HSCT on 100 days mortality.

Methods:We retrospectively reviewed the medical charts of 331 patients who underwent HSCT at our institution from January, 2007 until December, 2013. Information on patients' body weight (BW) was measured daily, starting at admission. The highest BW recorded until until the first 10 days post-SCT (D+10) was used to calculate the BW increase in relation to the baseline BW. Based on our previous study, we used a cutoff of 6% gain in BW to identify a group of patients with increased risk of complications. The primary endpoint was mortality within 100 days post HSCT. Overall survival (OS) was estimated from the time of HSCT until death, and surviving patients were censored at last follow-up. A logistic regression model and a Cox model were fit to determine variables that predicted death within 100 days and OS, respectively. Statistical analysis was performed with STATA (v11.0) and alfa error was defined as 5%.

Results: Median age was 43 years old (range <1 year-76 years) and 60% were male. HSC sources included autologous (46%), matched related donors (16%), matched unrelated donors (12%), cord blood units (16%) and mismatched related/unrelated donors (10%). Diagnosis included acute leukemia or chronic myeloid disorders (34%), lymphoma/multiple myeloma (42%) and non-malignant hematological disorders (24%). Twenty-one percent of patients had a ≥6% gain in BW until the first 10 days post-HSCT. These patients had developed an increased inflammatory state after the start of the conditioning regimen: there was no difference in baseline levels of C-reactive protein (4.8 mg/L vs. 7.4, p=0.37), but by D+10 patients who gained more BW had higher CRP (116.7 mg/L vs. 178.6 mg/L; p=0.01). Patients with increased gain in BW until D+10 had a decreased OS (HR 2.33, p<0.0001, 95% CI 1.45-3.73). The mortality within 100 days was 47% in the increased BW vs. 17% in the control group (p<0.0001). Among 18 patients who died within 100 days and had a ≥6% BW gain by D+10, causes of death included pneumonia (N=4), septic shock (N=9), fungal endocarditis (N=1), ischemic stroke (N=1) and disease progression (N=3). In a logistic regression analysis, after adjusting for age, sex, diagnosis and type of SCT, a ≥6% BW gain by D10 was associated with an increased risk of being dead within 100 days (coefficient 1.75; p<0.0001; 95% CI 0.90-2.61). Similarly, in a multivariate Cox analysis after adjusting for age, sex, diagnosis and type of SCT, a ≥6% BW gain by D10 was an independent risk factor for survival (HR 2.72, p<0.0001, 95% CI 1.65-4.48). A landmark analysis at D+100 revealed that the negative impact of weight gain by D+10 on survival was restricted to the first 100 days, as after this time point there was no survival difference between the two groups (HR 1.35; p=0.41; 95% CI 0.65-2.83).

Conclusion: A ≥6% BW gain by D+10 is a risk factor for early mortality in both autologous and allogeneic HSCT. The most common cause of death in these patients is infectious-related complications. An increase in BW is related to the development of an inflammatory state, probably induced by the conditioning regimen. BW gain is a simple variable that can be easily used to determine prognosis of patients post-allogeneic HSCT, and further studies are needed to determine its etiology.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution