Hematologic and Renal Trends in Cats Undergoing Continuous Renal Replacement Therapy with Adjunctive Molidustat Administration: A Retrospective Case Series

1. Introduction

Acute kidney injury (AKI) in cats is associated with substantial morbidity and mortality [1,2], and timely diagnosis and aggressive management are critical determinants of outcome [3]. In clinical practice, AKI frequently occurs in cats with evidence of pre-existing chronic kidney disease [4,5,6], resulting in an unstable metabolic and hematologic environment. These patients are at increased risk of severe azotemia, uremic complications, and fluid and electrolyte derangements, which may necessitate advanced renal support [3].

In severely affected feline patients, particularly those with refractory uremia, fluid overload, or electrolyte imbalances, continuous renal replacement therapy (CRRT) has emerged as an important therapeutic option [4]. CRRT allows gradual solute and fluid removal and is often favored in hemodynamically unstable patients [3]. Although the use of CRRT in small animal medicine, including feline patients, has increased in recent years, published reports describing associated complications and longitudinal clinical outcomes in cats remain limited [3].

Anemia is a well-recognized complication of CRRT in both human and veterinary patients. The development of anemia during CRRT is multifactorial and has been attributed to repeated blood sampling, extracorporeal circuit–related blood loss, anticoagulation, inflammation, and disturbances in iron metabolism [7,8,9,10]. In cats, whose circulating blood volume is relatively small, these factors may impose a disproportionately greater hematologic burden during extracorporeal therapies [3]. Consequently, progressive anemia may develop despite transfusion support and standard anemia management strategies. Because the extracorporeal circuit volume used during CRRT may represent a substantial proportion of the total circulating blood volume in cats, hematologic responses during treatment may differ from those observed in larger species [3].

Management of anemia in critically ill patients undergoing CRRT typically relies on whole blood transfusions and, in some cases, erythropoiesis-stimulating agents (ESAs) [7,11]. While these interventions may transiently improve oxygen-carrying capacity, they are often insufficient to prevent ongoing anemia during prolonged or repeated CRRT [7]. In human critical care literature, anemia frequently persists or worsens in patients receiving CRRT despite transfusion and ESA therapy [7], underscoring the complex and multifactorial nature of anemia in this setting.

Recently, hypoxia-inducible factor prolyl hydroxylase inhibitors (HIF-PHIs) have been introduced as a novel therapeutic class for the management of anemia associated with chronic kidney disease [12]. Molidustat stimulates a dose-dependent erythropoietic response in healthy cats [13] and may be an alternative for the management of CKD-associated anemia in cats [14]. However, data regarding hematologic responses in cats undergoing CRRT for acute kidney injury, particularly in the presence of severe systemic illness and repeated extracorporeal interventions, are extremely limited [15].

CRRT represents a highly dynamic treatment environment characterized by continuous extracorporeal circulation, repeated interventions, and profound metabolic and inflammatory fluctuations [3,16]. Under these conditions, hematologic trends may differ substantially from those observed in stable CKD patients, and interpretation of anemia management responses is inherently challenging. Descriptive reporting of hematologic and renal trends in feline patients undergoing CRRT is therefore of clinical relevance.

The objective of this retrospective observational study was to describe temporal changes in hematologic and renal parameters in cats undergoing continuous renal replacement therapy for acute kidney injury with evidence of pre-existing chronic kidney disease, and to report clinical observations related to anemia management during this high-risk treatment period.

2. Materials and Methods

2.1. Study Design

This study was designed as a retrospective observational study. Medical records of client-owned cats treated with continuous renal replacement therapy (CRRT) at a single referral center were reviewed to evaluate temporal hematologic and renal trends during and after extracorporeal renal support.

2.2. Case Selection

Cats were eligible for inclusion if they met all of the following criteria: (1) underwent CRRT for the management of acute kidney injury with evidence of pre-existing chronic kidney disease; (2) had serial hematologic and renal laboratory data available before, during, and after CRRT; and (3) received standard-of-care clinical management for anemia during hospitalization. Three cats fulfilled these criteria and were included in the final analysis. Baseline characteristics and treatment overview for the three cats are summarized in Table 1.

2.3. CRRT Protocol

Continuous renal replacement therapy was performed using a double-lumen 8-Fr hemodialysis catheter placed in the right jugular vein for vascular access. Treatments were conducted using a Fresenius Multifiltrate System (Fresenius Medical Care AG & Co. KGaA, Bad Homburg, Germany) in pediatric continuous venovenous hemodialysis (CVVHD) mode with polysulfone dialyzers (Ultraflux AV paed, Fresenius Medical Care AG & Co. KGaA, Bad Homburg, Germany).

Because of the relatively small circulating blood volume of feline patients [17], the extracorporeal circuit and dialyzer, with an approximate priming volume of 60 mL, were primed using crossmatch-compatible whole blood prior to treatment initiation.

All systemic anticoagulant medications were discontinued during treatment, and anticoagulation of the extracorporeal circuit was achieved using a continuous-rate infusion of unfractionated heparin administered at 2.0–4.8 IU/h according to clinical requirements. Blood and dialysate flow rates were adjusted according to individual patient tolerance and clinical response throughout treatment sessions.

Treatment protocols were standardized across all cases. Case 1 underwent a single 6-hour session using intermittent hemodialysis (IHD). Case 2 underwent three sessions with durations of 5, 8, and 6 hours, respectively, consisting of one IHD session followed by two prolonged intermittent renal replacement therapy (PIRRT) sessions. Case 3 underwent three sessions with durations of 10, 8, and 6 hours, all performed using PIRRT. No episodes of filter clotting or clinically apparent extracorporeal circuit blood loss were observed during any treatment session. Based on the body weight of the included cats (3.5–3.8 kg), the extracorporeal circuit volume represented approximately 23–31% of the estimated circulating blood volume.

2.4. Data Collection

Clinical, hematologic, and biochemical data were extracted from electronic medical records. Hematologic variables included hematocrit (HCT), hemoglobin concentration, and absolute reticulocyte count. Renal variables included blood urea nitrogen (BUN), serum creatinine, and serum phosphorus concentrations.

Treatment-related variables recorded included the timing and number of CRRT sessions, timing and volume of whole blood transfusions, use of erythropoiesis-stimulating agents, and adjunctive administration of molidustat as part of clinical anemia management. All medications and interventions were administered at the discretion of the attending clinician as part of routine patient care.

Laboratory values were organized chronologically relative to CRRT initiation and subsequent treatment course to facilitate evaluation of temporal trends. No interventions were performed solely for the purpose of data collection or analysis.

2.5. Outcome Assessment

The primary outcomes of interest were temporal changes in hematologic and renal parameters during and after CRRT. Analyses focused on descriptive patterns of anemia progression, reticulocyte response, and renal parameter fluctuations rather than absolute numerical comparisons or inferential statistical testing.

2.6. Data Presentation and Analysis

Data were summarized descriptively and visualized using line graphs to illustrate individual patient trajectories over time. Given the small sample size and observational nature of the study, no inferential statistical analyses were performed.

During the preparation of this manuscript, the authors used Claude Opus 4.5 (Anthropic) to assist with data analysis and generation of the figure illustrating hematologic and renal trends (Figure 1). The authors have reviewed and edited the output and take full responsibility for the content of this publication.

Detailed individual clinical data, including CRRT indications, treatment timelines, transfusion history, erythropoietic support, and serial laboratory values, are provided in the Supplementary Material to support interpretation of the reported trends.

3. Results

3.1. Clinical Course

Case 1

An 8-year-old neutered female domestic shorthair cat presented with acute kidney injury and evidence of pre-existing chronic kidney disease. Due to rapidly worsening azotemia and persistent uremic signs refractory to medical management, a single session of continuous renal replacement therapy (CRRT) was performed, accompanied by a type A whole blood transfusion (40 mL).

Following CRRT, progressive anemia was observed, and a single dose of darbepoetin alfa was administered. Oral molidustat therapy was initiated during the post-CRRT period as part of clinical anemia management. Hematocrit declined during hospitalization, while an increase in absolute reticulocyte count was observed after discontinuation of CRRT. Blood urea nitrogen (BUN) and serum creatinine concentrations decreased markedly following CRRT and subsequently showed partial stabilization during follow-up.

Case 2

A 12-year-old neutered male Persian cat was referred for severe azotemia and uremia associated with acute-on-chronic kidney disease and partial ureteral obstruction. Continuous renal replacement therapy was initiated shortly after admission, and a total of three CRRT sessions were performed during hospitalization. Each CRRT session was accompanied by a concurrent type A whole blood transfusion (40 mL).

Progressive anemia developed despite transfusion support, and both darbepoetin alfa and oral molidustat were administered as adjunctive therapies. Hematocrit declined throughout hospitalization, whereas reticulocyte counts increased following completion of CRRT. Blood urea nitrogen and serum creatinine concentrations decreased after each treatment session but demonstrated intermittent rebound patterns consistent with dialysis-dependent renal failure.

Case 3

A 10-year-old neutered male domestic shorthair cat (3.5 kg) of feral origin presented in stupor with ACKD, uremic encephalopathy, and hyperammonemia. Concurrent systemic infection (FHV, FCV, and Mycoplasma spp. by PCR) with marked leukocytosis (WBC 43.5 K/µL) was present. Additional comorbidities included ascites, hepatic lipidosis, pancreatitis, and HCM stage B1. Three PIRRT sessions were performed (10 h on 2024.10.12, 8 h on 10.14, and 6 h on 10.18), each with concurrent type A whole blood transfusion (40 mL). BUN and creatinine fell from 130.5 and 6.97 mg/dL at admission to a nadir of approximately 48.0 and 1.54 mg/dL after the third session, with BUN rebounding above 140 mg/dL between sessions. Neurological status partially improved. Molidustat (2.5 mg/kg orally once daily) was initiated early; no ESA was administered. HCT declined from 39.9% to approximately 20.8%. The markedly elevated admission reticulocyte count (102.2 K/µL) likely reflected inflammatory erythropoiesis [18,19]; counts fell to 8.0–9.9 K/µL during CRRT before recovering modestly to 44.7 K/µL after the third session.

3.2. Hematologic and Renal Trends

In all three cats, hematocrit (HCT) showed a consistent downward trend during the CRRT period despite repeated whole blood transfusions (Figure 1A). Cat 2 (orange) and Cat 3 (green), both of which underwent three CRRT sessions, exhibited stepwise decreases in HCT following each session. Cat 1 (blue), which received a single CRRT session, also demonstrated a post-CRRT decline in HCT. Although transient stabilization or mild increases in HCT were observed immediately after transfusion, these effects were not sustained, and overall anemia progression was observed in all cases.

Renal parameters responded promptly to extracorporeal therapy. Blood urea nitrogen (BUN) and serum creatinine concentrations decreased markedly immediately following CRRT initiation in all cats (Figure 1B,C). In Cats 2 and 3, subsequent rebound increases in BUN and creatinine were observed during follow-up, consistent with ongoing uremic production in the setting of dialysis-dependent acute kidney injury. In contrast, Cat 1 showed more sustained post-CRRT stabilization of azotemia, suggesting partial renal recovery.

Absolute reticulocyte counts demonstrated interindividual variability (Figure 1D). In Cats 1 and 2, reticulocyte counts increased following CRRT discontinuation, indicating regenerative erythropoietic responses despite preceding critical illness and transfusion exposure. In Cat 2, this response was particularly pronounced following a single administration of darbepoetin alfa in combination with molidustat. Cat 3 showed only a mild reticulocyte increase, which may reflect the suppressive effects of concurrent systemic inflammation and infection. Peak hematologic and renal parameters and overall trends for each cat are summarized in Table 2.

4. Discussion

This retrospective case series describes temporal changes in hematologic and renal parameters in three cats undergoing continuous renal replacement therapy (CRRT) for acute kidney injury and explores the clinical course associated with concurrent administration of the hypoxia-inducible factor prolyl hydroxylase inhibitor (HIF-PHI) molidustat. A consistent finding across all cases was the progression of anemia during CRRT despite transfusion support, while evidence of erythropoietic recovery was observed in some cats following discontinuation of CRRT.

CRRT is an essential life-support modality for managing severe AKI in both human and veterinary medicine; however, anemia is a well-recognized complication of this therapy [4]. The pathogenesis of CRRT-associated anemia is multifactorial—including repeated blood sampling, extracorporeal circuit–related blood loss, anticoagulation, inflammatory responses, and disturbances in iron metabolism—based primarily on evidence from human critical care and canine renal replacement therapy studies [16]. Comparable feline-specific mechanistic studies remain scarce, although cats are particularly vulnerable to anemia during intensive care [20]. In the present study, the extracorporeal circuit volume represented approximately 23–31% of the estimated circulating blood volume of the included cats. Even though the circuit was primed with whole blood, this relatively large extracorporeal volume may have increased susceptibility to hematologic instability during renal replacement therapy. The consistent decline in hematocrit observed in all three cats in this series aligns with previous reports describing anemia in small animal continuous renal replacement therapy patients [4].

Despite repeated whole blood transfusions in all cases, hematocrit declined progressively during the CRRT period. This observation supports the concept that CRRT-associated anemia cannot be fully explained by inadequate transfusion support or erythropoietin deficiency alone [8,16]. This interpretation is further supported by the absence of filter clotting or clinically apparent extracorporeal circuit blood loss during treatment. Therefore, overt circuit-related blood loss alone is unlikely to explain the progressive decline in hematocrit observed in these cats, suggesting that multiple concurrent mechanisms contributed to anemia development.

Renal parameters demonstrated marked improvement immediately following CRRT but showed variable rebound patterns among individual cats. In two cases, intermittent increases in blood urea nitrogen and creatinine following initial improvement were consistent with dialysis-dependent acute kidney injury, characterized by ongoing uremic generation in the setting of intermittent extracorporeal clearance [21,22]. These fluctuations emphasize that hematologic trends observed during CRRT occur within a highly unstable metabolic and inflammatory environment rather than a steady physiologic state.

Notably, increases in absolute reticulocyte counts were observed in some cats after completion of CRRT, indicating that erythropoietic activity was not completely suppressed during treatment and may recover as inhibitory factors diminish. Differences in reticulocyte response among cases likely reflect interindividual variability in disease severity, inflammatory burden, and adjunctive therapies. In particular, the modest reticulocyte response observed in the cat with concurrent systemic inflammation may reflect inflammation-mediated suppression of erythropoiesis [18,19], whereas the more pronounced response in another cat may have been influenced by concurrent ESA administration, despite a similarly high CRRT burden.

At the time of treatment, molidustat was used for the management of nonregenerative anemia associated with CKD in cats [23], but not specifically for use in ACKD or during CRRT; therefore, its use in all three cases was off-label. The decision to administer molidustat was based on the presence of underlying CKD-associated anemia and the clinician’s judgement that potential benefits in supporting erythropoiesis outweighed the risks in these critically ill cats.

Although anemia progressed during active CRRT, no abrupt worsening or clinically significant adverse events temporally associated with molidustat administration were observed. Following discontinuation of CRRT, stabilization of hematocrit and evidence of erythropoietic activity were noted in some cats. However, this study was not designed to evaluate the efficacy of molidustat, and no conclusions regarding its contribution to hematologic recovery can be drawn. Observed reticulocyte responses cannot be clearly separated from physiologic recovery following CRRT or from the effects of transfusion and ESA use.

Overall, these findings provide descriptive insight into the hematologic course of cats undergoing CRRT and highlight the predictable progression of anemia despite transfusion support. The observations underscore the need for future prospective studies to better define anemia pathophysiology and to systematically evaluate anemia management strategies in feline patients receiving CRRT. Clinicians should anticipate progressive anemia during feline CRRT and monitor hematologic parameters proactively, even in the presence of transfusion support.

4.1. Limitations

This study has several important limitations. First, the small number of included cats and the retrospective observational design preclude statistical analysis and causal inference. Findings are descriptive in nature and should not be generalized beyond similar clinical contexts. This limitation reflects the inherent challenges of studying critically ill small animal populations undergoing CRRT [21,22].

Second, the absence of a control group prevents differentiation of the individual effects of CRRT, transfusion, ESA use, and adjunctive molidustat administration. Consequently, it cannot be determined whether observed hematologic changes were attributable to specific interventions or to post-CRRT physiologic recovery. Similar limitations have been reported in both veterinary and human CRRT studies [7,21].

Third, variables that may influence anemia development and erythropoietic response, including anticoagulation protocols, blood sampling frequency, inflammatory status, and iron metabolism indices, were not systematically assessed due to limitations of retrospective medical records. The absence of these data limits mechanistic interpretation of the observed trends [24]. Furthermore, mechanistic biomarkers, including serum iron concentration, ferritin, transferrin saturation, endogenous erythropoietin concentration, hepcidin, and inflammatory cytokines, were not available [8,9]. Consequently, the mechanisms proposed in this study should be regarded as hypothesis-generating rather than definitive.

Finally, the cats included represented a heterogeneous population of critically ill patients with varying comorbidities and degrees of renal dysfunction. Accordingly, results should be interpreted within the context of similarly severe clinical presentations rather than extrapolated to all feline patients receiving CRRT. Nevertheless, serial hematologic observations in feline patients undergoing CRRT remain exceedingly scarce, and these findings provide preliminary data that may inform the design of future prospective studies.

5. Conclusions

This retrospective observational study describes temporal hematologic and renal trends in cats undergoing continuous renal replacement therapy for severe acute kidney injury with evidence of pre-existing chronic kidney disease. Despite repeated whole blood transfusions, progressive anemia occurred consistently during the CRRT period, while evidence of erythropoietic recovery was observed in some cats following discontinuation of CRRT.

These findings emphasize the complex and multifactorial nature of anemia in feline patients receiving CRRT and suggest that transfusion support alone may be insufficient to prevent anemia progression in this high-risk clinical setting. The descriptive data presented herein provide clinically relevant insight into anemia progression and management challenges during feline CRRT and may inform the design of future prospective studies. To our knowledge, this study represents one of the first reports describing serial hematologic trajectories during CRRT in cats and provides preliminary observational data to support future investigations into CRRT-associated anemia in feline patients.