A Systematic Literature Review to Assess Long-Term Outcomes Associated with Splenectomy in Patients with Chronic Immune Thrombocytopenia

Introduction: Chronic immune thrombocytopenia (cITP) is an autoimmune disorder defined by low platelet count (<100 x 10⁹/L) lasting ≥12 months in the absence of other causes of thrombocytopenia. Splenectomy is an option for patients with cITP who fail to respond to oral corticosteroids and/or intravenous immunoglobulin or relapse after treatment is discontinued. A systematic literature review (SLR) conducted in 2004 (Kojouri et al) identified articles describing outcomes associated with splenectomy in patients with cITP. The objective of this study was to update this SLR with a focus on contemporary data on long-term outcomes (≥12 months of follow-up).

Methods: MEDLINE, Embase, Cochrane CENTRAL and recent congresses were searched in June 2018. Results were screened against predefined criteria by two independent researchers. Included studies assessed patients with cITP (N≥15) who underwent splenectomy; studies of patients with secondary ITP, newly diagnosed ITP, and/or persistent ITP were excluded unless separate outcomes were reported for cITP subgroups. Outcomes of interest were clinical efficacy (response and relapse rates), safety (rates of complications), mortality, and health-related quality of life (HRQoL). Prospective or retrospective clinical studies or real-world study types were included. English-language studies published during or after 2000 were included, with no geographic restrictions.

Results: The literature search identified 3140 records for title-abstract screening. Of these, 159 full-text studies were evaluated and 108 were included in the analysis. Most studies (93) were retrospective. Fifteen prospective studies (9 interventional, 6 observational) but no randomized controlled trials were identified. Nine studies were comparative (all retrospective): splenectomy vs rituximab (3), splenectomy vs rituximab vs romiplostim (1), and splenectomy vs non-splenectomy (5).

Reports of the long-term efficacy of splenectomy varied widely, with multiple definitions of response and remission across the heterogeneous study types. Among 40 studies, the mean complete response (CR) rate within 12 months of surgery was 77% (median: 81%; range: 26-97%). Relapse rates varied widely, ranging from 0-94% among 47 studies with ≥12 months of follow up. Five of 7 studies reporting remission rates at multiple time points at ≥1 year noted a decrease in clinical remission over time. Mortality generally increased with length of follow up: in studies with ≤1 month of follow-up (28 studies) the mean mortality rate was 1% (range: 0-5%), while in studies with 1-5 years of follow-up (20 studies) and ≥5 years of follow-up (15 studies), the mean mortality rate was 2% (range: 0-17%) and 11% (range: 0-30%), respectively. Four studies reported that long-term response rates were higher with splenectomy than rituximab; all other efficacy comparisons were inconclusive.

Although 11 of 15 prospective studies and 61 of 93 retrospective studies reported some safety information, there were very limited data on the long-term safety of splenectomy. Commonly reported complications were bleeding (mean: 14%; median: 12% range: 0-50%; 22 studies), infections (mean: 8%; median: 4% range: 0-33%; 38 studies), venous thromboembolism (VTE) (mean: 5%; median: 3% range: 0-21%; 27 studies) and sepsis/septic shock (mean: 2%; median: 0%; range: 0-11%; 18 studies). Rates of postoperative complications (≤30 days) ranged from 3-50% (mean: 13%; 31 studies), and 2 studies suggested that older age may be associated with higher rates of postoperative complications. HRQoL data were rarely reported (3 studies).

Rates of remission, relapse, and infections for studies reporting at least 1 of these outcomes at 1 or more discrete time points are shown in Figure 1.

Conclusions: Although more than 100 studies reported long-term outcomes for patients with cITP treated with splenectomy, available evidence on the durability of response and long term safety are limited. In general, most measures of efficacy declined over time, while complications (infections, bleeding, VTE) and mortality increased over time. The extent to which the outcomes for splenectomy differ from currently available treatments is unclear. Additional data are needed to understand the long-term benefits and risks of splenectomy in patients with cITP.

View largeDownload slide

View largeDownload slide

Close modal