Introduction

Thrombotic thrombocytopenic purpura (TTP) is a rare hematologic disorder characterized by excessive platelet activation due to ADAMTS13 deficiency, leading to microvascular thrombosis [1]. The majority of TTP patients have autoimmune TTP, from an acquired inhibitor of ADAMTS13, although inherited deficiency of ADAMTS13 can occur. Therapeutic plasma exchange (TPE) has been standard of care for acute TTP, along with adjunctive immune suppression with corticosteroids and rituximab for autoimmune TTP [2–5]. The addition of caplacizumab in acute TTP and active surveillance in remission to detect impending disease relapse can also improve outcomes [3,6–9]. Strategies are needed to improve acute TTP care, and to aggressively prevent TTP relapses in order to minimize the cumulative risk of death and disability. [10–13].

Although scoring systems such as PLASMIC and the French score can facilitate TTP diagnosis, scores have reduced utility in some populations, such as older individuals. [14,15]. The similarities between TTP and other thrombotic microangiopathies (can result in diagnostic uncertainty and delayed treatment, with resultant morbidity and mortality [16–18]. Many knowledge gaps and barriers exist to ideal TTP clinical care, and care is heterogeneous [19]. Knowledge gaps include the optimal duration of TPE, the use of a TPE taper, rituximab dose intensity, and frequency of surveillance testing to detect biochemical ADAMTS13 relapse before overt clinical relapse [2,9,20–22]. TTP care requires coordination across medical services (intensivists, proceduralists, hematology, blood bank/laboratory medicine, and often nephrology) and the inpatient-to-outpatient transition of care, which may lead to care delays and gaps in care. Furthermore, the TPE procedure may be conducted by hematology, transfusion medicine, or nephrology depending on local practices, potentially amplifying the clinical consequences of the knowledge gaps. Clinical pathways can improve care by structuring and standardizing multidisciplinary care for specific clinical problems [23,24].

To address these and other gaps in TTP care observed in our health system that contributed to poor outcomes, we formed a TMA Team as reported by Gordon et al., and subsequently implemented a TTP Pathway developed by the TMA Team [25]. We herein report our experience with the TTP Pathway at our center. The TTP Pathway is a multi-disciplinary effort to facilitate care and coordination, and establish evidence-based treatment and surveillance for TTP.

Methods

This prospective cohort study was conceptualized with the DMAIC (define, measure, analyze, improve, control) methodology of Six Sigma [26–28]. Detailed methods are found in the supplement (S1 File). WVU Medicine is a hospital system providing care to patients in rural Appalachia within West Virginia, western Maryland, eastern Ohio, and southwestern Pennsylvania.

Results

Fifty-eight patients were identified by ADAMTS13 testing. Thirty-two patients were excluded for not having TTP. Non-TTP etiologies were: disseminated intravascular coagulation in the setting of malignancy, infection, or vasculitis (N = 11), infection or sepsis (N = 12), missing data (N = 3), lupus or vasculitis (N = 2), drug induced thrombocytopenia (N = 1), cirrhosis (N = 1), other TMA (N = 1), and history of TTP without acute TTP in study window (N = 1). The remaining 26 patients with at least 1 acute TTP episode in the study period were included for analysis: 13 patients received usual care, 16 patients received TTP Pathway care; and 3 patients received both usual care and at time of relapse were subsequently treated on the TTP Pathway. Patient characteristics are listed in Table 1. TTP patients were predominantly women, white race, and the majority had a history of previous TTP before the study period. Socioeconomic and geographic barriers to care were common. There were more new TTP episodes in the Pathway cohort, yet the difference was not significant (95% CI −0.28 to 0.36, P = 0.781). The majority of patients (72.5%) were in the bottom third of affluence for the nation by area deprivation index (ADI). There were two cases of COVID-19 associated TTP. All patients had autoimmune TTP, and no cases of congenital TTP were identified.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Demographic data.

https://doi.org/10.1371/journal.pone.0325417.t001

Usual care and TTP Pathway care were notably different (Table 1). The use of rituximab in acute TTP was increased with the TTP Pathway (95% CI −0.36 to −0.88, P = 0.003), as was use of caplacizumab (95% CI −0.27 to −0.81, P = 0.001). However, caplacizumab was not available for 3 of the 4 years in the study before the TTP Pathway began. The number of TPE sessions was significantly higher on the TTP pathway (median 6 vs 12, P = 0.0011 by two sample Wilcoxon rank sum test), despite the frequent use of caplacizumab that enabled shorter TPE duration. Limiting analysis to non-caplacizumab treated episodes, TPE duration was longer with TTP Pathway care (median 5.5 vs 15, P = 0.027 by two sample Wilcoxon rank sum test). All patients on the TTP Pathway received rituximab during the acute TTP episode, except 1 patient that died with acute TTP while receiving daily TPE (Table 1). There was no difference between usual care and TTP Pathway care in: corticosteroid use, presentation at the academic center versus community hospitals (95% CI −0.40 to 0.28, P = 0.721), having had TPE (95% CI −0.18 to 0.056, P = 0.31), or leaving hospital against medical advice (95% CI −0.037 to 0.29, P = 0.14).

The primary outcome of TTP-RRD at 90 days occurred in 69% of patients with usual care versus 6% of patients on the TTP pathway (95% CI 0.35 to 0.90, P = 0.0004) (Table 2). The relative risk of TTP-RRD at 90 days was higher with usual care (RR 9.0, 95% CI 1.28 to 63.02, P = 0.006), and the NNT for benefit on the TTP Pathway was 1.59. The TTP-RRD relative risk reduction of 91.3% met the pre-specified outcome for pathway success (50%); TTP-RRD absolute risk reduction was 63%. The TTP Pathway attained the secondary objective of increased use of rituximab for acute TTP (RR 3, 95% CI 1.43 to 6.27, P = 0.0006). Four patients died of acute TTP, three from the usual care cohort (S1 Table). TTP-related death was reduced (RR 0.27, 95% CI 0.032 to 2.3), but not significantly different between usual care and TTP pathway care. TTP relapse at 90 days was reduced with TTP pathway care (62% vs 0%, 95% CI 0.36 to 0.88, P = 0.0002; RR 0, 95% CI not estimable due to zero events, P = 0.002), and the 90 day relapse NNT was 2.8. During biochemical surveillance on the TTP Pathway, 31% of patients were preemptively treated with rituximab due to declining ADAMTS13 activity with the intent to prevent TTP relapse, no patients were treated with preemptive rituximab before the 90 day analysis period. No patients in the usual care group had ADAMTS13 surveillance or preemptive rituximab therapy (95% CI −0.083 to −0.54, P = 0.015). TTP Pathway care also significantly reduced the primary outcome of TTP-RRD over the entire study period (Fig 1, log-rank P = 0.0018). The HR for TTP-RRD with usual care during the project was 12.58 (95% CI 1.61 to 98.67, P = 0.016) compared to TTP Pathway care. Median distance from the academic center to patient was 82 miles, and 85% of patients were more than 50 miles from the academic center. Distance >50 miles from the center was not associated with TTP-RRD at 90 days in the usual care group (95% CI −0.27 to 0.75, P = 0.38), however numbers were small. Newly diagnosed TTP can have increased mortality as a confounding factor, and more new TTP cases were seen on the TTP Pathway [35,36]. In subgroup analysis, the proportion of newly diagnosed TTP patients with TTP-RRD was higher with usual care (0.60 vs 0.14, 95% CI 0.04 to −0.96, P = 0.095) but the difference was not statistically significant given small numbers for comparison (RR 4.2, CI 0.60 to 29.5, P = 0.22; log-rank P = 0.25) (S2 Fig).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. TTP Pathway Outcomes.

https://doi.org/10.1371/journal.pone.0325417.t002

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Primary outcome of TTP-RRD.

Kaplan-Meier analysis of all patients with acute TTP treated with usual care or TTP Pathway care during the study period, with long-rank testing indicated.

https://doi.org/10.1371/journal.pone.0325417.g001

Outpatient TTP management site of care differed with the TTP Pathway. Academic center follow-up with classical hematology was increased (58% vs 93%, 95% CI −0.66 to −0.42, P = 0.031; RR 1.6, 95% CI 0.97 to 2.63), and fewer surviving patients had no outpatient follow up after the acute TTP episode (25% vs 0%, 95% CI 0.005 to 0.49, P = 0.04; RR 0, 95% CI not estimable due to zero events). Duration of caplacizumab therapy with academic center follow-up was 1 month after hospital discharge for all TTP Pathway patients based on ADAMTS13 recovery, except for the patient who declined follow-up: that patient was treated locally for 3 months and outpatient ADAMTS13 data were not available. On the TTP pathway, 50% of patients treated with caplacizumab experienced some adverse event attributed to caplacizumab including: new onset heavy menses (N = 2); severe needle phobia with injection anxiety (N = 2, required caregiver to administer injections), mild epistaxis (N = 1); one patient had “profuse” bleeding at the site of femoral venous access 4 hours after temporary apheresis catheter removal with a platelet count of 275, this required pressure and a dressing change. No adverse event required caplacizumab discontinuation.

Caplacizumab availability and use at regional sites was explored for acute therapy of clinically suspected TTP before patient transfer to the academic center for TPE. However, due to low utilization and cost concerns at regional sites this was ultimately not feasible. During the project the use of caplacizumab became more common over time, as a result of ISTH guideline recommendations [2]. As part of the Control process, we did observe that inpatient physicians were often using caplacizumab with the standard anticipated TPE duration of 12–14 sessions, which was not intended. Pathway clarification on TPE duration was attained with the TMA Team pharmacist who rounded with the consult service. Other difficulties were limited clinical follow-up appointment availability for care continuity due to limited providers in classical hematology, onerous prior authorizations for caplacizumab, initial difficulty obtaining caplacizumab (adding to formulary, pharmacy availability), delay in obtaining a hematology pharmacist for the team, and TMA Team clinical effort was not directly supported. At our institution medications given at home (caplacizumab) need authorization by the prescribing physician, which was an additional burden on the consult team until the addition of the pharmacist.

Discussion

Acute TTP is a rare and life-threatening medical emergency with substantial morbidity for survivors [12,13,37]. TTP care can be difficult to coordinate because of the number of medical services involved and knowledge gaps that affect clinical decisions. Using a multidisciplinary approach to TTP care with a TMA Team, we developed a TTP Pathway by identifying areas for improvement in usual care. The TTP Pathway was associated with significant and clinically meaningful improvements in patient care, such as a 91% relative reduction in TTP-RRD at 90 days, a NNT of 1.59 to prevent one episode of TTP-RRD at 90 days, and routine use of rituximab in acute TTP that was later recommended in guidelines [2]. The negative clinical outcome of TTP-RRD was markedly reduced with the TTP Pathway (RR 9.0 with usual care). Over the duration of the project a marked reduction of TTP-RRD was observed (HR 12.58 with usual care). Although TTP mortality was meaningfully reduced with the Pathway (23% vs 6%, RR 0.27), despite more initial TTP episodes in this cohort where mortality can be higher, this improvement was not statistically different from usual care with our small patient numbers [35,36].

Novel TTP treatments (caplacizumab) can improve TTP outcomes, but more effective use of existing treatments (TPE, immune suppression) and care coordination (TMA Team, ADAMTS13 surveillance) can also be beneficial [2,3,7,25,38,39]. Improving care delivery could be valuable in both high and low resource settings, and care pathways can help ensure use of guideline-based care. [17,19] Although TMA Teams are found at many academic centers, there is an absence of data showing the effectiveness of a TMA Team, except for one report on non-TTP TMA [40]. Our TTP Pathway streamlined care and demonstrated clear clinical benefit of the TMA Team, but barriers to care were noted during the project.

Our catchment area is rural and underserved. In this study 72.5% of TTP patients are in the bottom third of affluence in the nation by ADI. Sub-specialist classical hematology and TPE are available at only one site in the region. We observed that use of caplacizumab greatly facilitated TTP transitions of care and outpatient management, where many patients had difficulty traveling for TPE. The ISTH TTP guideline recommends routine use of caplacizumab in acute TTP, however this was controversial due to concerns about hemorrhage and cost [41,42]. Caplacizumab is expensive and was not cost-effective on analysis [43]. However, routine and early use of caplacizumab reduced TPE refractory TTP, can shorten hospital stays, can shorten TPE, and can be cost saving compared to delayed caplacizumab use or non-use [7,43,44]. We believe that routine caplacizumab use can specifically benefit rural and underserved patients who face disproportionate barriers to outpatient TPE, while also reducing relapses when used with upfront immune suppression. [44] The possibility of TPE-free TTP treatment would represent a considerable therapeutic advance, especially for rural and underserved patients where TPE is a disproportionate barrier to care. Although there are initial reports on TPE-free treatment, further studies are needed to demonstrate clinical effectiveness and cost effectiveness before this would become a standard treatment [45].

The ideal duration of TPE is not known. The benefit of a TPE taper is uncertain. Despite these knowledge gaps taper use was common in the USA and Canada in the pre-caplacizumab era, and many TPE cessation strategies were reported [46,47]. TTP guidelines either do not comment on TPE duration or taper, or recommend TPE until 2–3 days after platelet normalization, yet studies specifically analyzing TPE cessation strategies are lacking [2,5,48]. ADAMTS13 recovery after immune suppression and TPE may take 1–2 weeks or longer [3,44]. Longer duration TPE can function as a bridge in this window via antibody depletion. Indeed, others observed a significant delay in ADAMTS13 recovery with caplacizumab, attributed to less TPE and rituximab use [49,50]. The high TTP recurrence rate of 38% in the HERCULES trial within 30 days for non-caplacizumab treated patients supports this idea. The short reported TPE duration (median 7 days), infrequent rituximab use (48%) and high recurrence rates were similar to our observations for usual care, which prompted this improvement project [7]. Our usual care TPE duration of 6.5 days was shorter than other studies (median 10 in Coppo et al.; median 15 and 17 in Gomez-Segui et al.; mean 13 in Radhwi et al; median 11 in Page et al.), but similar to some studies (median 7 Scully et al.; median 5 in Van de Louw, et al.) [7,37,51–53]. Given our prior experience (S1 File) and studies suggesting improved responses with more TPE or taper use, the TTP Pathway focused on either extending TPE in conjunction with rituximab immunosuppression or using caplacizumab in conjunction with rituximab immunosuppression [21,22,54]. Future studies on TPE length or tapers may be moot or not feasible with increasing use of upfront caplacizumab and rituximab as reported here and as in Coppo et al., except in low resource settings where cost concerns may be prohibitive [44].

This report shows that TTP outcomes can be meaningfully improved with a TMA Team and TTP Pathway. We show that the TMA Team provides demonstrable benefit in our health system. Similar approaches can improve outcomes at other centers for TTP. Our study has notable limitations. First, our TMA Team is not a formal or supported entity, and all inpatient therapy was at the discretion of the treating physicians (typically solid tumor oncologists). As such, care heterogeneity was expected, as observed with TPE tapers in patients treated with caplacizumab, where no taper was intended. Because of this, our study was neither designed nor intended to be a comprehensive trial on the total use and function of the TMA Team. Second, confounding did occur because of the multiple interventions used on the Pathway. This was by design with the DMAIC method, and is distinct from a randomized trial. The events and causal factor analysis determined that multiple interventions were required simultaneously to maximally reduce the critical to quality characteristic of TTP-RRD (S1 File), therefore we cannot determine the relative effectiveness of each intervention. Third, this Pathway was specific for TTP, and presence of another TMA would require a distinct care path. However, because of the nature of the intervention, the rarity of TTP cases, and small patient numbers utilizing the TTP Pathway, the TTP Pathway was much easier to maintain than a similar intervention for hemophilia patients, where direct clinical involvement was substantial and ultimately not sustainable [55]. Finally, TTP is a rare disease and as expected in a single-institution setting for a rare disease our patient numbers are small, although this is the first report of a TMA Team changing TTP outcomes to our knowledge.

Conclusions

The use of a TTP Pathway incorporating evidence-based care, coordination between specialties, and routine surveillance improved TTP care and reduced relapses. Importantly, TTP Pathway care addressed distinct disease-specific needs during acute management in hospital and in the ambulatory clinic. This Pathway was easy to maintain after development and required only minimal clinical effort for continuation. Using a quality improvement approach to address barriers to care can improve outcomes in TTP, and for other hematologic disorders at other centers. The TMA Team has an important role in ensuring that TMA clinical care continues to improve as knowledge and therapies evolve.

Supporting information

Acknowledgments

The authors wish to thank colleagues in oncology, pulmonology, obstetrics, nephrology, and laboratory medicine for participating in clinical care.

Generative AI statement

Artificial intelligence was not used for any portion of this work.

Open Practices Statement

The dataset supporting this study is included in the supplementary files (S1 Table). All data required to replicate the results of the study are available in the provided manuscript tables and supplement, with the exception of data determined to be protected health information by the policies of the West Virginia University Office of Human Research Protections (886 Chestnut Ridge Road, Morgantown WV, 25406). Ethical and legal restrictions prevent sharing individual de-identified patient data including patient age, race, dates of therapy, ADI, and zip code because these are potentially identifying of individual persons in this study. These restrictions were imposed by the WVU Institutional Review Board, and Office of Human Research Protections pursuant to Data Protection Certificate #1084 for this project. Requests for protected health information can be made by contacting the IRB administrator at irb@mail.wvu.edu.