High incidence of suicidal ideation in a series of patients with sickle cell disease after hematopoietic stem cell transplantation

TO THE EDITOR:

Sickle cell disease (SCD) is a risk factor for reduced quality of life (QOL), low self-esteem, hopelessness, depression, anxiety, and cognitive impairment.^1,2 Estimates of lifetime prevalence of clinical depression in patients with SCD range from 22% to 50% vs overall 13.2% in the United States.^1,3 Chronic pain is in a bidirectional relationship with depression, with up to 85% of pain sufferers developing depression⁴ and depression increasing the risk of developing pain⁵ and chronic pain increasing the rate of death by suicide.⁶ Suicidal ideation (SI) and suicide attempts (SAs) are noteworthy phenomena even among patients diagnosed with depression. In general population, estimates of 1-year incidence of SI for younger adults are 6.6% among 18- to 25-year-olds and 4.0% among 26- to 49-year-olds.⁷ Among people with major depressive disorder, the 1-year incidence rises to 26.3%.⁷ Depression in SCD is understudied. One study found a 29% lifetime prevalence of SI and 8% lifetime prevalence of SAs among patients with SCD.⁸ Depression in SCD correlates with poorer adherence, increased days of pain,⁹ severity of pain, emergency room visits, and transfusions.^1,2,10 

Hematopoietic cell transplantation (HCT) is the only cure for SCD.¹¹ In adult patients, stable engraftment and reversal of SCD symptoms are achieved in over 90% of patients receiving matched related donor grafts, with success depending on variables including age, donor type, and conditioning regimen.¹¹ There are very few studies about QOL and none about clinical depression or suicidality in patients with SCD undergoing HCT. However, patients with cancer undergoing HCT report poor QOL, depression, and SI and have an increased rate of death by suicide for years following transplant.^12-14 We noted a frequent clinical complaint of SI in our adult patients with SCD and present this case series of SI after HCT.

The Columbia University Institutional Review Board approved this retrospective chart review. Electronic medical records (EMRs) were reviewed to identify all patients diagnosed with SCD who underwent HCT at the Columbia University Medical Center from 2014 to July 2018. We collected demographics, basic transplant data, duration of the initial stay, number of readmissions, and readmission diagnoses from the EMRs. Two psychiatrists (S.M.D. and E.J.L.) independently reviewed all relevant EMRs for 1 year after transplantation. All psychiatry notes explicitly included suicide assessments as part of the standard of care. All episodes of SI or behavior, as evidenced by patient reports, patient-observed behavior, or family report, were identified by our research team. Each patient record was classified by the presence or absence of SI and SA and by the number of days of SI. Each identified episode of SI was classified utilizing the Columbia Classification Algorithm of Suicide Assessment (C-CASA), which categorizes levels of suicide risk and has been found to have excellent inter-rater reliability.¹⁵ The C-CASA was applied retrospectively in chart review to classify the intensity of all SI or SA events.^15,16 

We assessed the incidence of SI in patients with SCD after HCT, compared these rates with their histories of depression, and assessed the demographic and clinical differences between suicidal and nonsuicidal groups. We analyzed by the Fisher exact test (categorical covariates) and Wilcoxon rank sum (continuous variables).

Of the 14 adult patients who were included in the study, 7 were men and 7 were women, 8 were of African or Afro-Caribbean descent, and 6 were of Hispanic descent. Seven patients received matched related donor cells, 6 unrelated donor cells, and 1 had a genetically modified autologous transplant. Successful stable engraftment defined by chimerism and the absence of vaso-occlusive crises was achieved in all 14 patients and 0 had chronic graft-versus-host disease. Hemoglobin S levels decreased either to 0% or equilibrated with the Hemoglobin S trait level of the donor in all patients. Length of stay ranged from 32 to 71 days.

The patients underwent psychosocial clearance by a clinical social worker (14) and a psychiatrist (13). Nine patients had a clinical history of depression, SI, or both. We categorized the patients into binary groups of any history of depression or SI or no history. All 14 patients were psychiatrically stable, with neither SI nor acute major depressive episode at the time of transplant; 4 were on antidepressants. All patients with a psychiatric history were actively followed up by a team psychiatrist (A.D.M.) and all patients had continuous access to psychiatric care (Table 1).

Six patients (42.9%) reported an SI and 1 had an SA within 1 year after HCT, and all patients who reported an SI had a C-CASA classification of “suicidal ideation.” None of the patients died or engaged in other self-injury. There was no relationship between suicidality and gender, age, race, or donor type (Table 1). Of the 5 patients with no history of depression, SI, or SA, none (0%) had an SI in the year after transplantation. Of the 9 patients with any history of depression, SI, or SA, 6 (66.7%) expressed an SI after transplantation. The association between history of depression, SI, or SA with posttransplant SI was highly significant (odds ratio = ∞; P = .0310). Having a history of depression or suicidality before transplantation was associated with an increased number of total pain-related readmissions after transplantation (mean [range], 1.333 [0-3] vs 0 [0-0]; P = .0304) (Table 2).

We found a 43% rate of SI in our case series of patients with SCD in the post-HCT year across age, race, gender, and donor type, despite 100% engraftment. This is higher than that reported in studies on the SI incidence for either SCD or depression.⁷ There are no published data on SI rate in post-HCT SCD. Contributions may include severe SCD correlating with opting for HCT, worse pain, increased medically related stress immediately after-transplant, and the isolation required during HCT. Given that chronic pain continues for more than a year after transplantation,¹⁷ a longer follow-up might reveal an initial period of increased suicidality followed by improvement.¹⁸ It concerns us greatly that some patients psychiatrically deteriorated after HCT. Having a history of depression or suicidality correlated with pain readmissions. Our findings raise the question of whether patients are sufficiently emotionally prepared for the post-HCT year, during which pain and depression are likely to continue.

This 14-patient case series was exploratory. The sample size was too small to compare by age, disease burden, and other factors that may be critical. Additional limitations include the retrospective chart review design and the lack of quantitative psychometric scales at evaluation. A retrospective review methodology increases the risk of undercounting events and the total rate of SI might even be higher. We had no available control group and a comparable psychiatric follow-up for non-HCT patients was not uniformly available. There may be unique characteristics about our program, or the patients it attracts and accepts, given the presence of a dedicated niche psychiatrist, which further decreases its generalizability. Strengths included the duration of follow-up, integrated team, thorough review, the inclusion of inpatient and outpatient documentation, outcome-blinded psychiatrist raters, and the inclusion of a diverse population of patients.

There is insufficient research in SCD supporting specific screening, management, and monitoring practices, potentially exacerbating disparities in care.¹⁹ Given longstanding inequities in clinical care and biomedical research for people of African and Afro-Caribbean descent, patients with SCD, and especially those facing multiple layers of inequities, we consider ongoing work in this area to be vital. Our findings require formal prospective consideration for the evaluation of major depression, SI, and SA in post-HCT SCD. We provisionally recommend thorough, universal, and required screening for depression and SI in patients with SCD anticipating transplant and that transplant programs have psychiatrists available for ongoing support.

Contribution: A.D.M. designed the research and wrote the manuscript; S.G.C. managed and analyzed the data; A.H. participated in background research and wrote the manuscript; E.J.L. and S.M.D. collected the data; and M.Y.M. participated in data analysis and interpretation, and wrote the paper.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Adrienne Mishkin, Department of Psychiatry, Columbia University Medical Center, 622 W 168th St, PH Center, New York, NY 10032; e-mail: adm2172@cumc.columbia.edu.