Haptoglobin Administration for Intravascular Hemolysis: A Systematic Review

Intravascular hemolysis is the rupture of red blood cell (RBC) in the vascular vessels with subsequent release of their content in the blood stream. Hemolysis is associated with a variety of adverse consequences, linked to impaction of ruptured RBC membranes and to the release of hemoglobin in the blood (plasma-free hemoglobin [pfHb]) [1]. The effects of pfHb are pleiotropic and pro-oxidative and include acute kidney injury (AKI) due to direct cytotoxic effects and tubular pfHb casts accumulation, vascular dysfunction due to nitric oxide pathway inhibition with resulting unbalanced vasoconstriction and impaired end-organ perfusion, platelet activation, and procoagulant effects with increased risk of thrombosis [1‒3].

Hemolysis may be caused both by intrinsic causes (chiefly hematologic processes resulting in RBC structural defects) and extrinsic causes, leading to direct RBC damage (including, but not limited to: autoantibodies, sepsis, trauma, burns, and mechanical circulatory support [MCS] pumps). While removal of the cause is paramount to resolve hemolysis, this is not possible in critically ill patients supported by life-saving temporary MCS. In this case, the blood-machine interface may generate and maintain the hemolytic process with possible end-organ detrimental consequences due to AKI, inflammation, and thrombosis in the already critically ill patient [4].

Physiologic clearance of pfHb is mediated by haptoglobin, a protein with high affinity to pfHb; the pfHb-haptoglobin complex is then cleared by the reticuloendothelial system [5]. However, when hemolysis is intense, the body haptoglobin is completely depleted, and then the excess pfHb may exert its noxious effects.

Therefore, exogenous haptoglobin administration may enhance pfHb clearance during hemolytic condition and prevent complications related to pfHb circulation. Unfortunately, available data on haptoglobin administration to prevent hemolysis-related end-organ damage in humans are few and sparse. Accordingly, we performed a systematic review with meta-analysis to summarize available evidence on the use of haptoglobin as a treatment to prevent complications related to hemolysis of any cause.

This review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [6‒8]. The following PICOS framework was used. “P – Population”: all patients receiving haptoglobin for hemolysis treatment or prevention; “I – Intervention”: haptoglobin administration; “C – Comparison”: any control treatment; “O – Outcome”: primary and secondary outcomes as detailed below; “S – Study type”: randomized, prospective, and retrospective observational studies, case series and case reports. This systematic review and meta-analysis was registered on the International Prospective Register of Systematic Reviews (PROSPERO; Registration No. CRD42023390043).

Two independent researchers searched the databases PubMed, BioMed Central, Embase, and the Cochrane Central Register of Controlled Trials until August 1st, 2023, and identified the relevant articles at title/abstract level. The full-text of articles that passed the initial screening was retrieved for detailed assessment for inclusion in the final analysis. The complete search strategy is presented in the online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000539363). In addition, the reference lists of reviews and expert opinions on the subject were also searched to retrieve missing studies. Duplicates were removed after discussion by all the authors. Disagreements were resolved by consensus with the help of a senior author. Non-English language studies were excluded.

Studies were included in the qualitative synthesis if they described haptoglobin administration as treatment or prevention of hemolysis of any cause. Only randomized controlled trials (RCTs) and retrospective controlled studies reporting at least one of the outcomes of interest were included in the quantitative synthesis. Studies on pediatric patients, animal studies, preclinical studies, reviews, and expert opinions were excluded.

Data extraction was performed by two blind researchers, and all the primary and secondary outcomes were collected and entered into a 2016 Excel spreadsheet (Microsoft, Redmond, WA, USA). Outcomes were defined as per author definition of each individual study. We abstracted data on study design and sample size, patients’ age, study clinical setting, indications and trigger for haptoglobin administration, haptoglobin dose and timing of administration, control treatment (if any), serum levels of creatinine, haptoglobin and free hemoglobin, adverse events, incidence of AKI, and mortality. Data were abstracted from tables, figures, and text, as available.

The primary outcome was the effect of haptoglobin on pfHb after 1 h. Secondary outcomes included pfHb change at 24 h; in-hospital mortality; occurrence of AKI; serum haptoglobin concentration after haptoglobin administration at 1 and 24 h; any reported adverse event.

Review Manager (RevMan) version 5.4.1 (Review Manager, The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen) and the “meta” package with RStudio (v. 1.3.1093) were used for quantitative synthesis of the haptoglobin’s effect size among studies comparing it against control. We used Cochran’s Q test and I² statistic to assess the heterogeneity of the results. I² values of less than 25% indicate low heterogeneity, 25–50% moderate heterogeneity, and greater than 50% high heterogeneity. A fixed-effects model was used. The Mantel-Haenszel statistical method was used to calculate individual and pooled odd ratios (OR) and corresponding 95% confidence intervals (CIs) for dichotomous outcomes. For continuous outcomes, we calculated the mean difference (MD) and corresponding 95% CIs using the inverse variance method. Meta-regression analyses were also performed, using select covariates of interest. A p value <0.05 was considered statistically significant.

The risk of bias of the included RCTs was assessed according to the Cochrane Handbook for Systematic Reviews of Interventions and using version 2 of the Cochrane risk-of-bias tool for randomized trials (RoB 2) [9]. We assessed the five items separately, and we rated the potential risk of bias as “low,” “some concern,” or “high” for each study. Any disagreement was resolved by consensus. Funnel plots are not reported as number of included studies was <10 in all analyses [10]. Quality assessment for observational studies included in this review was done with the Newcastle-Ottawa Scale (NOS) tool [11, 12].

A total of 573 articles were screened and, after detailed examination, a total of 13 articles were finally included in the review [13‒26] (consort diagram in online suppl. Fig. 1). A detailed description of the population, inclusion criteria, haptoglobin dosing, and outcomes is reported in Table 1 for the studies eventually included in the quantitative synthesis and meta-analysis. A description of the reports included in this review but not in the quantitative synthesis and meta-analysis is available in Table 2.

The overall sample size included 677 patients, 358 of which (52.8%) received haptoglobin. Three studies were RCTs [13, 14, 21], and six were case reports/series with a sample size of <5 patients each. One manuscript included both randomized data and a non-randomized case series [13]. One retrospective study included a crude overall analysis and a propensity-matched analysis [22]. One case report and one RCT involved pregnant patients [18, 21]. The most common clinical setting was cardiac surgery with cardiopulmonary bypass (CPB, 4 studies, 532 [78.6%] patients) [17, 19, 22, 23]. Bias risk analysis is reported in online supplementary Figures 2 and 3.

In most cases, an initial i.v. bolus of a fixed dose of 2,000 or 4,000 international units (IU) was administered (median [IQR]: 4,000 [2,000, 4,000] IU i.v. bolus). In the RCTs, initial bolus dose ranged from 2,000 to 4,000 UI [13, 14, 21]. A total of 5 manuscripts reported additional boluses (from one additional dose up to six additional doses within 6 days after the initial dose) [13, 18, 20, 22, 24]. Only 1 case series administered haptoglobin as continuous infusion drip (dose ranging from 4,000 to 10,000 IU/day) [16]. The maximum amount of haptoglobin infused in the first 24 h was 16,000 IU [24]; the median maximum amount of haptoglobin infused in the first 24 h was 4,000 (3,500, 8,000) IU. The total administered dose of haptoglobin during the treatment ranged from 2,000 to 36,000 IU (Tables 1, 2).

Three RCTs presented data on pfHb at any timepoint [13, 14, 21]. Two RCTs [13, 14] reported the pfHb values after the hemolytic insult (range 105 ± 69 to 267 ± 15 mg/dL; pooled mean 147 ± 94 mg/dL), three RCT [13, 14, 21] 1 h after the haptoglobin infusion (range 105 ± 69 to 26 ± 30 mg/dL; pooled mean 84 ± 70 mg/dL) and two RCT [13, 21] 24 h after haptoglobin infusion (range 23 ± 17 to 0 ± 0 mg/dL; pooled mean 6 ± 13 mg/dL). Pooled analysis showed that haptoglobin administration, as compared to control group, was associated with a lower level of free hemoglobin after 1 h after infusion (MD −11.28; 95% CI −15.80 to −6.75; p < 0.001; I² = 85%; with three studies included). At 24 h, haptoglobin infusion was associated with a lower level of pfHb (MD −2.65; 95% CI −4.73 to −0.57; p = 0.001; I² = 92%; with two studies included). These findings are summarized in Figure 1a, b.

Mortality data were available in three studies (two RCTs [13, 21], 1 retrospective observational study [20]). Pooled analysis showed no difference in all-cause mortality between haptoglobin-treated patients and control group (OR 1.41; 95% CI: 0.49 to 4.95; p = 0.520; I² = 0%; with three studies included). These findings are summarized in Figure 2a. These results were maintained when a +0.5 correction factor was applied to the event counts of both arms of the study by Yamamoto et al. [21] (OR 1.38; 95% CI: 0.49–3.82; p = 0.535; I² = 0%).

Haptoglobin serum levels 1 h following haptoglobin infusion were reported by three RCTs [13, 14, 21], and haptoglobin serum levels 24 h following haptoglobin infusion were reported by three studies (2 RCTs [13, 21], 1 retrospective observational study [20]). Haptoglobin infusion, as compared to control group, led to higher serum concentration of haptoglobin at 1 h (MD 45.22; 95% CI: 34.45–56.00; p < 0.001; I² = 94%; with three studies included) and at 24 h (MD 20.67; 95% CI: 20.67–35.68; p < 0.001; I² = 97%; with three studies included). These findings are summarized in Figure 1c, d.

Data on haptoglobin effect upon renal function are available from three studies. Haptoglobin infusion, as compared to control group, was associated with lower incidence of AKI, by an OR 0.64; 95% CI: 0.44–0.93; p = 0.020; I² = 51%; with three studies included. These findings are summarized in Figure 2b.

No adverse events or side effects associated with haptoglobin use were reported, although dedicated discussion on drug safety was lacking in many studies.

Considering the wide time span of the included studies, a meta-regression analysis selecting study publication year as covariate was run: no significant association was found between study publication year and outcomes (online suppl. Fig. 4, 6, 8, 10, 12). A meta-regression analysis considering median haptoglobin dose infused as covariate was also run: no significant association was found between the haptoglobin dose used and outcomes (online suppl. Fig. 5, 7, 9, 11, 13).

The main findings of this study may be summarized as follows (Visual Abstract):

• 1.

  exogenous haptoglobin was associated with lower concentration of pfHb and with higher circulating haptoglobin concentration during hemolysis;

• 2.

  exogenous haptoglobin infusion was associated with lower risk of AKI during hemolysis;

• 3.

  no significant adverse events were reported during exogenous haptoglobin infusion during hemolysis caused by a large variety of etiologies.

In this systematic review and meta-analysis, we found that haptoglobin was administered in a wide variety of patients with hemolysis, including pregnant patients. Haptoglobin infusion was associated with subsequent reduction in pfHb plasma concentration and no increase in mortality or adverse events. To the best of our knowledge, this is the first systematic review assessing the role of haptoglobin as a pharmacologic agent to prevent and treat hemolysis-related complications. Notably, several preclinical studies found that haptoglobin infusion was able to blunt inflammatory response and limit kidney injury in a model of stored RBC packs related hemolysis [27]. Similarly, haptoglobin reduced the toxic effect of pfHb in a mouse model of sickle cell disease [28] and attenuated the effects of chronic exposure to pfHb on mouse’s lung vasculature [29]. Taken together, this evidence identifies, whatever the initial hemolysis trigger, a common effect of haptoglobin in mitigating the deleterious pleiotropic effects of pfHb.

The clinical data we summarized also suggest potential for human haptoglobin use to prevent or treat complications associated with hemolysis. In particular, a potential benefit of haptoglobin administration emerges in reducing the level of pfHb and preserving renal function in patients with hemolysis with little risk of adverse events. Of note, in the studies included, haptoglobin has also been administered in “vulnerable” populations (e.g., pregnant patients) frequently excluded from pharmacologic clinical trials in many settings. Based on the most common dosing regimen, two to three doses of 2,000–4,000 IU each within 24 h seem sufficient to reduce pfHb and potentially prevent AKI. Macroscopic signs of hemolysis have generally been used to trigger treatment, and disappearance of signs of hemolysis to interrupt it. The CPB-related hemolysis during cardiac surgery was the most investigated setting. Building on this, also patients receiving continuous-flow MCS for cardiogenic shock may benefit from treatment with haptoglobin as hemolysis is a frequently reported side effect of these devices [30, 31]. Specifically, micro-axial flow pumps and VA-ECMO circuits are particularly prone to hemolysis development that may occur in a percentage as high as 38% or higher, and especially in the early phase of support [30, 32]; very limited therapeutic options are available for such patients. In these patients, intense shear forces act upon RBCs, causing membrane cell distortion and rupture. The resulting hemolysis may be intense and exceed the scavenging capacity of the body with resulting high levels of circulating pfHb. In addition, liver failure complicating CS may cause further haptoglobin deficiency in this setting.

Although the clinical use of haptoglobin is, to our knowledge, limited to burns, trauma, and CPB settings, CS patients treated with short-term MCS who develop severe hemolysis may also benefit from haptoglobin replenishment. Therefore, we call for further investigations focused on this population, to clarify the effect of increasing haptoglobin concentration by means of exogenous haptoglobin infusion in limiting the end-organ damage caused by the hemolysis generated by MCS shear forces. Currently, however, haptoglobin is not commercially available in Europe or in the USA, and we hope that this systematic review may renew the interest in this pharmacologic approach that represents a potential option for these patients.

Our study has some limitations. There is a large heterogeneity in the hemolysis etiology and age of participants of included studies. Several studies were old, with only one study published in the last 5 years. Most studies were non-randomized and lacked a control group. Included RCTs were old and with a small sample size. In addition, heterogeneity was high for several outcomes: clinical settings, baseline pfHb levels, and drug dosing regimen were highly variable across studies, likely contributing to this result. However, this remains the most comprehensive systematic review on the topic, and our study should be considered preliminary rather than confirmative.

Data from our study support designing future interventional studies investigating the role of haptoglobin as a pharmacologic agent to prevent and treat hemolysis-related complications. Future studies should confirm that exogenous haptoglobin administration is associated with lower pfHb levels and improved renal function, and that these changes translate into clinically significant improvement in outcomes. The most attractive settings for future studies seem patients receiving CPB or MCS. Two doses of 2,000 or 4,000 IUs within 24 h seem adequate to exert a potential clinical benefit with limited adverse events.

Haptoglobin administration has been used in patients with hemolysis from any cause to treat or prevent hemolysis-associated adverse events. Haptoglobin may reduce levels of pfHb and preserve kidney function without an increase in adverse events.

Ethical approval and consent were not required as this study was based on publicly available data.

The authors have no conflicts of interest to declare.

The Authors thank Fondazione Rodolfo Ferrari for funding the publication of this study.

Study design and final approval of the manuscript: L.B., R.L., A.B., and A.M.S.; implementation: A.D.-F.; statistical plan: Y.K. and S.F.; data collection/curation: F.C., E.F., and M.P.; data analysis: S.A., B.P., R.L., and L.B.; and manuscript: B.P. and L.B.

PROSPERO Registration No. CRD42023390043.

All data analyzed during this study are included in this article and its online supplementary material files. Further inquiries can be directed to the corresponding author.