Introduction: Hyperviscosity syndrome (HVS) is a clinical feature associated with Waldenström macroglobulinemia (WM). The overproduction of an IgM paraprotein can increase serum viscosity and induce signs and symptoms of HVS, prompting the need for WM-directed therapy. Current consensus panel guidelines recommend initiating treatment only for symptomatic HVS rather than a specified serum IgM level (Kyle et al, 2003). However, many clinicians treat for an elevated IgM in the absence of hyperviscosity-related findings to pre-empt HVS and any associated sequelae. Empiric treatment for WM patients with high serum IgM levels has been proposed as a reasonable criterion for treatment initiation regardless of symptomatic status due to the risk of HVS (Treon, 2015). We therefore sought to determine the serum IgM level threshold for which the risk of HVS would be supportive of treatment initiation.

Methods: We identified 825 untreated patients who met the consensus diagnosis for WM (Owen et al, 2003), and who received care in the WM clinic at our Institution between January 1999 and June 2016. The occurrence of symptomatic HVS was then determined, and serum IgM levels at the time of HVS diagnosis were stratified. For patients with an IgM >=3,000 mg/dL, a stratified hazard analysis was performed to compare the risk of HVS associated with increasing serum IgM levels. The outcome of interest was hazard ratio (HR) with 95% confidence interval (CI). P-values <0.05 were considered statistically significant.

Results: Among 825 untreated patients, 112 (14%) cases of HVS were identified. The incidence of HVS in patients with serum IgM levels between 3000-3999, 4000-4999, 5000-5999, 6000-6999 and >=7000 mg/dl was 3%, 22%, 30%, 61% and 79%, respectively. The HR of HVS in patients with serum IgM levels 4000-4999, 5000-5999, 6000-6999 and >=7000 mg/dl was 7.5 (95% CI 3.4-16.8), 9.8 (95% CI 4.7-20.4), 28.2 (95% CI 13.2-60.3) and 26.8 (95% CI 12.5-57.7), respectively (IgM 3000-3999 was baseline; Figure 1A). When compared to patients with a serum IgM 3,000-5,999 mg/dL, patients with a serum IgM >=6,000 mg/dL had a 4.9-fold increased risk for development of HVS (HR 4.87, 95% CI 3.06-7.56; Figure 1B). No symptomatic HVS was observed among patients with a serum IgM <3,000 mg/dL. HVS patients with a serum IgM level >=6,000 mg/dL were significantly more anemic and thrombocytopenic, as well as more likely to have a serum B2M >3.0 mg/dL. HVS patients with a serum IgM >=6,000 mg/dL were also more likely to receive emergent plasmapheresis versus those HVS patients with a serum IgM 3,000-5,999 mg/dL (77% vs. 58%; p=0.04). All but one patient received WM-directed intervention in response to symptomatic HVS; one patient refused treatment.

Conclusion: Patients with a serum IgM >=6,000 mg/dL are at a significantly increased risk for developing symptomatic HVS. Given the magnitude of the risk increase described herein for symptomatic HVS, a serum IgM >=6,000 mg/dL may reasonably be considered as a criterion for initiation of WM-directed therapy.


Ghobrial:Amgen: Honoraria; Noxxon: Honoraria; BMS: Honoraria, Research Funding; Celgene: Honoraria, Research Funding; Novartis: Honoraria; Takeda: Honoraria. Castillo:Otsuka: Consultancy; Abbvie: Research Funding; Biogen: Consultancy; Janssen: Honoraria; Millennium: Research Funding; Pharmacyclics: Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.