Hepatorenal and Full Blood Count Profiles of Stroke Survivors Compared With Controls: A Case–Control Study

ABSTRACT

Stroke is not merely a cerebrovascular event but a systemic disorder with multi‐organ implications, including the liver and kidneys. Despite this growing body of knowledge, significant gaps persist. Most studies characterizing poststroke renal, hepatic, and hematological profiles originate from high‐income settings, with limited data from sub‐Saharan African populations who face distinct genetic, environmental, and healthcare access challenges. A case–control study was conducted at the Tamale Teaching Hospital between March and September 2025. The study involved 69 participants, aged 31–78 years, of whom 20 (29.0%) were stroke survivors (cases). Of the 69 participants, 39 (56.5%) were female. A single venous blood sample was collected and analyzed for full blood count, liver and renal function variables. Bivariate analysis revealed no significant differences in liver and renal function; however, the mean cell volume (MCV [fL]), the red cell distribution width (RDW‐CV [%]), eosinophils (%), and plateletcrit (PCT [%]) were lower, whereas the mean cell hemoglobin concentration (MCHC [g/dL]) was higher in the stroke group. After adjusting for confounders, including body mass index (BMI) and occupation, the indirect bilirubin level was 0.143 mmol/L higher in the stroke group. In addition, the MCHC was 0.398 g/dL higher, while the hematocrit (HCT), MCV, and RDW‐CV were lower in the stroke group by 0.206 (%), 0.121 (fL), and 0.514 (%), respectively. This study suggests that stroke survivors exhibit distinct hematological profiles, characterized by alterations in RBC morphology and possibly low‐grade hemolysis or altered bilirubin metabolism independent of major liver or renal dysfunction. These changes may reflect the persistent systemic inflammation, oxidative stress, and erythropoietic adaptation following a stroke.

In a case–control study of 69 participants (20 stroke survivors, 49 controls; aged 31–78 years) at Tamale Teaching Hospital, Ghana (March–September 2025), venous blood was analyzed for full blood count, liver and renal function. After adjusting for BMI and occupation, stroke survivors showed higher indirect bilirubin (0.143 mmol/L) and MCHC (0.398 g/dL), and lower hematocrit (−0.206%), MCV (−0.121 fL) and RDW‐CV (−0.514%) than controls, with no significant liver or renal differences, indicating distinct red‐cell morphology and mild hyperbilirubinemia after stroke.

1. Introduction

Stroke remains one of the leading causes of mortality and long‐term disability among adults globally, with low‐ and middle‐income countries (LMICs) disproportionately burdened, particularly in sub‐Saharan Africa (SSA) [1]. Stroke is a major public health concern in Ghana, characterized by high incidence, prevalence, and case‐fatality rates, placing a significant strain on healthcare systems and socio‐economic structures [2]. The pathophysiological cascade of stroke includes systemic inflammatory response and oxidative stress that can extend beyond the central nervous system, potentially affecting secondary organ function [3].

Emerging pieces of evidence suggest that stroke is not merely a cerebrovascular event but a systemic disorder with multi‐organ implications. Hepatic and renal function are critically implicated due to their central roles in detoxification, metabolism, inflammatory modulation, and drug clearance [4, 5]. Poststroke liver dysfunction may arise from hemodynamic changes, systemic inflammation, or as a consequence of polypharmacy during treatment and rehabilitation [6]. Similarly, acute kidney injury (AKI) and chronic kidney disease (CKD) are both independent risk factors for and common complications following stroke, sharing connected pathways involving neurohormonal activation, endothelial dysfunction, and oxidative stress [7]. Moreover, hematological indices, indicators of oxidative stress, systemic inflammation, and erythrocyte integrity, are increasingly recognized as biomarkers for the severity, risk, and prognosis of stroke [8].

Despite this growing body of knowledge, significant gaps persist. Most studies characterizing poststroke renal, hepatic, and hematological profiles originate from high‐income settings, with limited data from SSA populations who face distinct genetic, environmental, and healthcare access challenges [9]. Furthermore, studies among stroke patients in Ghana are mostly cross‐sectional or surveys, which are less robust, unlike case–control studies, which compare disparities between stroke survivors and nonstroke individuals recruited from similar hospitals, controlling for healthcare‐seeking behavior and baseline comorbidities [10, 11]. Understanding these biochemical and hematological disparities in stroke patients is crucial for holistic patient management, guiding and informing rehabilitation strategies tailored to the unique needs of stroke survivors in resource‐limited settings.

Therefore, this hospital‐based case–control study aimed to identify disparities in hepatic function, renal function, and full blood count variables between stroke survivors undergoing treatment and rehabilitation and non‐stroke control subjects recruited from the same tertiary hospital environment in Northern Ghana. By adjusting for key confounders such as body mass index (BMI) and occupation, the study sought to elucidate specific biochemical and hematological alterations associated with stroke survivorship in this understudied population.

2. Materials and Methods

2.1. Study Design and Setting

This case–control study was conducted at the Tamale Teaching Hospital (TTH) between March and September 2025. The TTH is a tertiary‐level healthcare facility that serves as a referral healthcare facility in northern Ghana and parts of neighboring Burkina Faso. The TTH has many specialized units and departments, including neurology, physiotherapy, imaging, and medical laboratory. All these units or departments have state‐of‐the‐art equipment and board‐certified personnel for the diagnosis, treatment, and rehabilitation of stroke and stroke patients. Persons who suffer from a suspected stroke event are first assessed by a general physician or other qualified healthcare personnel before being confirmed by a neurologist. Initial assessment of the patient includes imaging and laboratory investigations, including lipids and other tests. Upon successful treatment, the patient may be referred to the physiotherapy unit for rehabilitation to restore mobility and other locomotor functions.

2.2. Study Population and Sampling

The study involved 69 participants aged between 31 and 78 years. The cases were 20 (29.0%) clinically diagnosed stroke patients who were undergoing treatment and/or rehabilitation at the hospital during the study period. Controls consisted of 49 participants without a history of stroke, including patient relatives and nonstroke patients attending the hospital for conditions unrelated to cardiometabolic or neurological disorders. Of the 69 participants, 39 (56.5%) were female, while the rest were male. Stroke diagnosis was based on clinical assessment by a board‐certified neurologist, supported by neuroimaging findings documented in patients' medical records. Stroke was diagnosed by the acute onset of focal neurological deficits attributable to a cerebrovascular cause, confirmed by urgent neuroimaging. Clinically, patients who presented with sudden weakness or numbness, speech or language disturbance, visual loss, gait instability, or altered consciousness were included. Imaging tests conducted to confirm stroke included magnetic resonance imaging (MRI), computed tomography (CT), or magnetic resonance (MR) angiography. In addition, electrocardiography (ECG) and echocardiography may also be performed to assess cardioembolic sources, and carotid imaging to evaluate extracranial stenosis. The stroke could then be classified as ischemic or hemorrhagic, with etiological subtyping guiding acute treatment and secondary prevention. Laboratory tests are also usually requested, including blood glucose, full blood count, electrolytes, renal function, and coagulation profile. However, this study did not classify stroke into types to maintain statistical power. Adults aged ≥ 30 years with a confirmed diagnosis of stroke and who provided informed consent were included. The controls were adults without a prior history of stroke or transient ischemic attack. Participants with known active infections, malignancy, or those on lipid‐altering medications before the stroke event and pregnant women were excluded from the study. The study adopted a convenience sampling technique because this was a single‐center study where the number of stroke patients was fewer.

2.3. Variables

The outcome variable was stroke (stroke or nonstroke). Independent variables included socio‐demographic factors (sex, education, occupation, religion, smoking, and alcohol intake), anthropometric factors (age, height, weight, and BMI), hepatic factors (total protein, albumin, globulins, aspartate aminotransferase [AST], alanine aminotransferase [ALT], alkaline phosphatase [ALP], gamma‐glutamyl‐transferase [GGT], total bilirubin, direct bilirubin, and indirect bilirubin), renal factors (creatinine, urea, uric acid, potassium, sodium, and chloride), and hematological factors (red blood cell [RBC], hemoglobin [HGB], mean cell volume [MCV], mean cell hemoglobin concentration [MCHC], mean cell hemoglobin [MCH], hematocrit [HCT], red cell distribution width‐coefficient of variation [RDW‐CV], white blood cell count [WBC], neutrophils, lymphocytes, eosinophils, monocytes, basophils, platelets, and plateletcrit [PCT]). The socio‐demographic and anthropometric factors were considered as possible sources of confounding.

2.4. Data Sources and Measurement

A semi‐structured questionnaire was used to collect sociodemographic, anthropometric, and clinical information, including age, sex, occupation, BMI, and time since stroke onset (for cases). The participants were prepared on first contact before blood sample collection. A single venous blood sample was collected from each participant and dispensed into EDTA anticoagulant and gel‐separator vacutainer tubes using standard phlebotomy guidelines. The blood in the EDTA vacutainer tubes was mixed thoroughly with the anticoagulant to attain homogeneity. The anticoagulated blood samples were analyzed for a full blood count using an automated hematology analyzer within 2 h of collection. The blood in the gel‐separator tubes was allowed to clot before centrifugation at 3000 rpm for 10 min to obtain serum within 2 h after collection. The serum samples were then aliquoted into Eppendorf tubes in duplicate and stored at −25°C for later analysis. The frozen serum samples were removed from the freezer and allowed to thaw at room temperature without refreezing. The thawed serum samples were mixed thoroughly to attain homogeneity before being analyzed for liver and renal function factors using an automated biochemistry analyzer. All analyses were conducted in the hospital's accredited medical laboratory following internal quality control procedures.

2.5. Bias Mitigation

The selection of both cases and controls from the same hospital minimized, but may not have eliminated, all bias. In addition, the control group, recruited from the same hospital, may not be representative of the general population from which the stroke cases arose. They may vary slightly in their health‐seeking behavior, socioeconomic status, or comorbid conditions, potentially leading to an underestimation or overestimation of associations. Although the a priori power calculation, anchored to a previously reported difference in serum uric acid between stroke patients and controls, indicated that 11 cases and 21 controls would provide 80% power at a two‐tailed α of 5%, this calculation was based on a single variable and does not guarantee adequate power for the full panel of hepatorenal and hematological parameters examined, particularly with only 20 stroke cases enrolled. The modest case sample size limits statistical power to detect associations for markers with smaller effect sizes, increases the risk of Type II error, and is reflected in the relatively wide confidence intervals around some adjusted estimates. The laboratory analyses were performed using standardized automated methods, reducing the likelihood of differential measurement error for hematological and biochemical variables. The processing and storage protocols were consistent for all samples, minimizing non‐differential misclassification. However, data on sociodemographic and lifestyle factors such as smoking and alcohol consumption were collected via a tool and were subject to recall and desirability bias. While the analysis adjusted for identified confounders (BMI and occupation), residual confounding remains possible. Important clinical confounders known to influence both stroke risk and the studied laboratory parameters, such as the specific subtype and chronicity of stroke, hypertension status, diabetes mellitus, dyslipidemia, cardiovascular disease, and detailed medication history, were not accounted for in the reported analysis. Unmeasured or inadequately measured confounding could distort the observed associations. The findings are derived from a single‐center study and may not be generalizable to other populations with different genetic backgrounds, lifestyles, or healthcare environments beyond the study setting in northern Ghana. While methodological efforts were made, including multivariable analyses that controlled for potential confounders, potential selection and unmeasured confounding biases should be considered when interpreting the results of this study.

2.6. Study Size

In a previous study, the mean serum uric acid level in stroke patients was 6.4 ± 1.21 mg/dL, and in controls was 5.10 ± 0.97 mg/dL [12]. Using a control‐to‐case ratio of 2.0, 80% power, and an α value of 5%, power analysis of a two‐tailed independent samples test returned 11 cases and 21 controls as the recommended minimum sample sizes.

2.7. Quantitative Variables

Continuous variables, including age, anthropometric indices, hepatic function parameters, renal function indices, and hematological parameters, were initially assessed for completeness, plausibility, and distributional characteristics. Normality of continuous variables was evaluated using formal normality testing, alongside screening for extreme outliers. Variables approximating a normal distribution were summarized as means with standard deviations, while non‐normally distributed variables were summarized using medians with interquartile ranges, as appropriate. For inferential analyses, quantitative variables were primarily analyzed as continuous measures to preserve statistical power and avoid information loss, consistent with current epidemiological best practices. No arbitrary categorization of continuous biochemical or hematological parameters was performed. To account for potential confounding, multivariable regression models were fitted with quantitative outcome variables entered in their continuous form. Key covariates, including BMI and occupation, were selected a priori based on biological plausibility and evidence from prior literature.

2.8. Statistical Methods

Data were entered into a Microsoft Excel Spreadsheet and analyzed using IBM SPSS Statistics for Windows, Version 27.0 (Armonk, NY: IBM Corp). Continuous variables were tested for normality using the Kolmogorov–Smirnov test and visual inspection of Q‐Q plots. Extreme outliers were identified using a box‐and‐whisker plot and assessed for potential data‐entry errors or biological plausibility. Extreme outliers and missing values (less than 5% of the total) were replaced using the subgroup series mean. Descriptive statistics are presented as means ± standard deviations or medians (interquartile ranges) for continuous variables, and frequencies (percentages) for categorical variables. Initial bivariate comparisons between case and control groups were performed using independent samples t‐tests for parametric data, Mann–Whitney U tests for nonparametric data, and chi‐squared (or Fisher's exact) tests for categorical variables. Univariable regression analysis was performed to determine associations. To control for potential confounding, variables showing a baseline association, specifically BMI and occupation, were included as covariates in multivariable regression models. The regression models were used to estimate adjusted odds ratios comparing stroke survivors with controls, with 95% confidence intervals. A two‐tailed p‐value < 0.05 was considered statistically significant.

2.9. Ethical Approval

Ethical approval for the study was obtained from the institutional ethics review committee of the University for Development Studies (UDS/RB/0086/25). Additionally, permission to conduct the study was granted by the management of TTH. The study followed the guidelines regarding human subject studies as outlined in the 1964 Declaration of Helsinki or later amendments. Written informed consent was obtained from all participants before enrollment, and the confidentiality of participant information was strictly maintained throughout the study.

3. Results

3.1. Participants

In all 77 persons were eligible for the study, comprising 23 stroke patients and 54 nonstroke controls. However, 73 persons, comprising 22 stroke patients and 51 nonstroke controls, were actually assessed for eligibility. Finally, 20 stroke patients (cases) and 49 nonstroke patients (controls), totaling 69 participants, were enrolled in the study because the rest had declined participation. The participants were not followed, and all 69 participants were included in the final analyses.

3.2. Socio‐Demographic and Anthropometric Attributes of the Study Population

The association analyses between stroke and sociodemographic factors are presented in Table 1. The results showed that males were about five times more likely to be diagnosed with stroke in the study population (OR = 4.812 [95% CI: 1.559–14.860]). Similarly, salaried workers were less likely to suffer a stroke and survive than unemployed participants (OR = 0.147 [95% CI: 0.036–0.600]). The anthropometric characteristics are summarized in Table 2. The body weight (kg) and the BMI (kg/m²) were higher in the nonstroke control group than in the stroke group.

TABLE 1.

The sociodemographic characteristics of the study population.

Variables	Category	Stroke	Control	Χ 2	p	OR (95% CI)	p
Sex	Female	6 (30.0)	33 (67.3)	8.061	0.007	1
	Male	14 (70.0)	16 (32.7)			4.812 (1.559–14.860)	0.006
Education	None	2 (10.0)	6 (12.2)	0.858	0.835	1
	Primary	1 (5.0)	5 (10.2)			0.600 (0.041–8.732)	0.708
	Secondary	7 (35.0)	13 (26.5)			1.615 (0.255–10.226)	0.610
	Tertiary	10 (50.0)	25 (51.0)			1.200 (0.206–6.977)	0.839
Occupation	Unemployed	7 (35.0)	7 (14.3)	11.387	0.003	1
	Self‐employed	8 (40.0)	8 (16.3)			1.000 (0.238–4.198)	> 0.999
	Salary	5 (25.0)	34 (69.4)			0.147 (0.036–0.600)	0.008
Religion	Islam	13 (65.0)	29 (59.2)	0.202	0.788	1
	Christian	7 (35.0)	20 (40.8)			0.781 (0.265–2.302)	0.654
Smoking	No	20 (100)	48 (98.0)	0.414	> 0.99	n/a
	Yes	0 (0.0)	1 (2.0)			n/a
Alcohol	No	19 (95.0)	43 (87.8)	0.818	0.664	1
	Yes	1 (5.0)	6 (12.2)			0.377 (0.042–3.353)	0.382
Duration of stroke (months)	≤ 12	10 (50.0)	—	—	—	—	—
	> 12	10 (50.0)	—	—	—	—	—

Note: The data are summarized as frequency (proportions), and association tests include chi‐squared/Fisher and logistic regression.

TABLE 2.

The anthropometric characteristics of the study population.

Variables	Stroke	Control	p
Age (years)	56.1 ± 11.9	51.7 ± 9.8	0.118
Height (m)	1.7 ± 0.1	1.7 ± 0.1	0.779
Weight (kg)	66.7 ± 12.2	76.9 ± 14.4	0.007
BMI (kg/m2)	24.3 ± 4.5	27.8 ± 5.2	0.011

Note: The data are summarized as mean ± standard deviation.

3.3. Differences in Hepatorenal Variables

A summary of the bivariate analysis of the differences in hepatorenal factors between stroke and nonstroke groups is shown in Table 3. No significant differences in hepatorenal variables between the groups were found. The univariable logistic regression analysis showed no significant association between stroke and hepatorenal factors. However, after adjusting for significant confounders, including BMI and occupation, the indirect bilirubin level was 0.143 mmol/L higher in the stroke group than the non‐stroke group (AOR = 1.154 [95% CI: 1.003–1.328]), as shown in Table 4.

TABLE 3.

Disparities in hepatorenal variables between stroke and nonstroke groups.

Variables	Stroke	Control	p
Total bilirubin (mmol/L)	15.4 (7.7–19.2)	13.6 (10.1–17.0)	0.615
Direct bilirubin (mmol/L)	4.2 (2.3–6.4)	3.6 (2.7–5.2)	0.508
Indirect bilirubin (mmol/L)	8.5 (5.4–14.7)	8.9 (5.4–12.0)	0.639
ALT (IU/L)	23.8 (20.9–30.5)	22.8 (17.8–33.7)	0.890
AST (IU/L)	23.0 (19.1–28.4)	23.0 (17.8–29.3)	0.716
ALP (IU/L)	136.6 (112.7–171.8)	130.3 (107.7–157.2)	0.459
GGT (IU/L)	23.6 (19.7–43.3)	27.6 (19.3–39.6)	0.756
Total protein (g/L)	67.6 (65.4–70.7)	67.1 (63.3–69.7)	0.601
Albumin (g/L)	35.0 (38.0–41.5)	38.7 (36.3–41.1)	0.832
Globulins (g/L)	28.1 (25.5–32.2)	27.9 (24.6–32.9)	0.776
Creatinine (μmol/L)	68.4 (55.7–81.5)	71.5 (66.3–92.0)	0.290
Urea (mmol/L)	3.7 (2.9–5.9)	4.2 (3.4–5.8)	0.214
Uric acid (mmol/L)	281.5 (209.5–333.5)	242.6 (182.1–306.8)	0.435
Potassium (mmol/L)	4.3 ± 0.4	4.2 ± 0.5	0.256
Sodium (mmol/L)	141.7 ± 4.1	141.9 ± 4.2	0.857
Chloride (mmol/L)	101.5 ± 4.4	102.9 ± 4.7	0.250

Note: The data are summarized as mean ± standard deviation or median (interquartile range).

TABLE 4.

Association between stroke and hepatorenal variables.

Variables	B	OR (95% CI)	p	B	AOR (95% CI)	p
Total bilirubin (mmol/L)	0.047	1.048 (0.973–1.128)	0.216	0.065	1.067 (0.973–1.169)	0.166
Direct bilirubin (mmol/L)	0.107	1.113 (0.879–1.408)	0.374	0.104	1.110 (0.831–1.481)	0.480
Indirect bilirubin (mmol/L)	0.071	1.074 (0.979–1.177)	0.131	0.143	1.154 (1.003–1.328)	0.045
ALT (IU/L)	−0.007	0.993 (0.946–1.042)	0.778	−0.007	0.993 (0.933–1.056)	0.817
AST (IU/L)	0.013	1.013 (0.947–1.084)	0.703	0.051	1.053 (0.966–1.147)	0.242
ALP (IU/L)	0.005	1.005 (0.993–1.016)	0.424	0.007	1.007 (0.994–1.021)	0.290
GGT (IU/L)	−0.007	0.993 (0.961–1.027)	0.694	−0.011	0.989 (0.950–1.030)	0.602
Total protein (g/L)	0.012	1.012 (0.927–1.105)	0.785	−0.028	0.973 (0.870–1.087)	0.626
Albumin (g/L)	0.019	1.019 (0.893–1.162)	0.781	0.006	1.006 (0.852–1.188)	0.943
Globulins (g/L)	0.027	1.027 (0.923–1.144)	0.623	−0.003	0.997 (0.867–1.145)	0.961
Creatinine (μmol/L)	−0.022	0.978 (0.951–1.006)	0.125	−0.025	0.976 (0.943–1.010)	0.162
Urea (mmol/L)	−0.083	0.920 (0.675–1.255)	0.601	−0.181	0.835 (0.570–1.223)	0.354
Uric acid (mmol/L)	0.674	1.962 (0.616–6.246)	0.254	0.650	1.915 (0.472–7.768)	0.363
Potassium (mmol/L)	−0.012	0.988 (0.871–1.121)	0.855	−0.043	0.958 (0.817–1.123)	0.597
Sodium (mmol/L)	−0.070	0.932 (0.828–1.050)	0.248	−0.076	0.927 (0.804–1.068)	0.293

Note: Results are presented as regression coefficients (B), odds ratios (OR), and adjusted odds ratios (AOR).

3.4. Differences in Full Blood Count Variables

The bivariate comparisons between the groups are summarized in Table 5. The results indicate that the MCV (fL), RDW‐CV (%), eosinophil (%), and PCT (%) were higher, while MCHC (g/dL) was lower in the stroke group than the control group. After adjusting for confounders, including BMI and occupation, the indirect bilirubin level was 0.143 mmol/L higher in the stroke group. In addition, the MCHC was 0.398 g/dL higher, while the HCT, MCV, and RDW‐CV were lower in the stroke group by 0.206 (%), 0.121 (fL), and 0.514 (%), respectively (Table 6).

TABLE 5.

Disparities in hematological variables between stroke and nonstroke groups.

Variables	Stroke	Control	p
HGB (g/dL)	12.8 ± 1.7	12.2 ± 2.1	0.260
HCT (%)	34.3 ± 5.3	37.2 ± 6.8	0.087
RBC (×106/μL)	4.7 ± 0.7	4.5 ± 0.6	0.356
MCV (fL)	73.6 ± 8.3	82.3 ± 11.8	0.004
MCH (pg)	27.5 ± 3.5	26.9 ± 3.8	0.559
MCHC (g/dL)	37.5 ± 3.9	32.9 ± 4.8	< 0.001
RDW‐CV (%)	9.9 ± 1.0	11.9 ± 3.1	0.005
WBC (×103/μL)	4.0 ± 1.1	3.4 ± 1.2	0.062
Neutrophil (%)	47.5 (43.1–53.6)	45.8 (41.9–53.4)	0.526
Lymphocytes (%)	47.8 (38.1–55.6)	50.3 (42.4–55.9)	0.324
Monocytes (%)	1.9 (0.9–4.8)	2.2 (0.3–4.6)	0.721
Basophil (%)	0.1 (0.0–0.3)	0.1 (0.0–02)	0.941
Eosinophil (%)	0.02 (0.01–0.22)	0.09 (0.00–0.21)	0.041
Platelets (×103/μL)	143.5 (128.0–207.0)	187.0 (153.0–217.0)	0.097
PCT (%)	0.1 (0.1–0.2)	0.2 (0.1–0.2)	0.005

Note: The data are summarized as mean ± standard deviation or median (interquartile range).

TABLE 6.

Association between stroke and full blood count variables.

Variables	B	OR (95% CI)	p	B	AOR (95% CI)	p
HGB (g/dL)	0.170	1.185 (0.882–1.592)	0.259	−0.031	0.970 (0.663–1.418)	0.873
HCT (%)	−0.074	0.929 (0.852–1.013)	0.094	−0.206	0.814 (0.710–0.933)	0.003
RBC (×106/μL)	0.389	1.475 (0.649–3.356)	0.354	−0.140	0.869 (0.292–2.590)	0.801
MCV (fL)	−0.071	0.932 (0.885–0.981)	0.007	−0.121	0.886 (0.819–0.959)	0.003
MCH (pg)	0.045	1.046 (0.901–1.215)	0.554	0.036	1.036 (0.856–1.255)	0.714
MCHC (g/dL)	0.208	1.231 (1.086–1.395)	0.001	0.398	1.489 (1.159–1.913)	0.002
RDW‐CV (%)	−0.418	0.658 (0.481–0.902)	0.009	−0.514	0.598 (0.400–0.893)	0.012
WBC (×103/μL)	0.436	1.546 (0.969–2.465)	0.067	0.410	1.506 (0.823–2.755)	0.184
Neutrophil (%)	0.013	1.014 (0.970–1.059)	0.546	0.022	1.023 (0.963–1.086)	0.462
Lymphocytes (%)	−0.019	0.981 (0.945–1.019)	0.330	−0.026	0.975 (0.928–1.023)	0.301
Monocytes (%)	0.004	1.004 (0876–1.152)	0.949	0.010	1.010 (0.873–1.169)	0.895
Basophil (%)	0.386	1.471 (0.203–10.683)	0.703	1.497	4.470 (0.335–59.735)	0.258
Eosinophil (%)	−1.911	0.148 (0.017–1.290)	0.084	−2.366	0.094 (0.004–2.035)	0.132
Platelets (×103/μL)	−0.008	0.992 (0.982–1.002)	0.109	−0.006	0.994 (0.984–1.005)	0.289
PCT (%)	−13.722	0.000 (0.000–0.106)	0.019	−9.404	0.000 (0.000–1.616)	0.062

Note: Results are presented as regression coefficients (B), odds ratios (OR), and adjusted odds ratios (AOR).

4. Discussion

This case–control study aimed to examine the disparities in hematological, hepatic, and renal function profiles between stroke survivors and nonstroke controls. After adjusting for confounders, including BMI and occupation, the indirect bilirubin level was 0.143 mmol/L higher in the stroke group. In addition, the MCHC was 0.398 g/dL higher, while the HCT, MCV, and RDW‐CV were lower in the stroke group by 0.206 (%), 0.121 (fL), and 0.514 (%), respectively.

The adjusted elevation in indirect bilirubin, a product of heme catabolism, was observed. This is consistent with recent evidence linking bilirubin metabolism to cerebrovascular disease. Indirect bilirubin is generated when heme oxygenase‐1 (HO‐1) catabolizes heme released from senescent or stressed erythrocytes; HO‐1 is markedly upregulated by oxidative stress and neuroinflammation following cerebral ischemia, so a modestly higher indirect bilirubin may represent a lingering marker of this antioxidant, cytoprotective response rather than overt hemolysis or hepatic dysfunction [13]. Clinically, total and direct bilirubin have been reported to correlate with stroke severity and, in several recent cohorts and meta‐analyses, with functional outcome and early neurological deterioration, although the direction of association is inconsistent across studies and appears to depend on the bilirubin fraction, timing of measurement and the population studied [14, 15, 16, 17]. This bidirectional evidence supports our interpretation that the modestly elevated indirect bilirubin observed here most plausibly reflects an adaptive antioxidant response to poststroke oxidative stress rather than pathological hyperbilirubinemia.

The alterations in RBC indices (higher MCHC, lower MCV, HCT, and RDW‐CV after adjustment) similarly find support in recent literature. Reduced hematocrit has recently been shown, in a large prospective registry, to independently predict poor functional outcome and higher mortality after ischemic stroke/TIA, reflecting a threshold relationship between red‐cell mass, blood viscosity/oxygen‐carrying capacity and cerebral perfusion [18]. RDW‐CV, a marker of anisocytosis driven by ineffective erythropoiesis, chronic inflammation and oxidative stress, remains one of the most consistently validated hematological prognostic markers in ischemic stroke, as confirmed by a 2024 meta‐analysis and meta‐regression pooling several thousand patients [19]. The attenuation and reversal of the RDW‐CV and MCHC associations after adjustment for BMI and occupation in our data underscore the strong confounding influence of nutritional and socioeconomic status on RBC morphology in this population. Beyond their diagnostic relevance, the hematological and hepatic parameters altered in this study have documented prognostic value in stroke populations. Elevated RDW‐CV has been independently associated with stroke severity, higher risk of stroke‐associated pneumonia, and both short‐ and long‐term mortality and poor functional outcome (modified Rankin Scale) after ischemic stroke, and continues to add incremental predictive value when combined with established clinical risk scores [13, 14, 19]. Reduced hematocrit similarly predicts a higher risk of poor 3‐month functional outcome and all‐cause mortality [18]. Serum bilirubin fractions, including indirect bilirubin, have been linked in recent meta‐analyses and large cohort studies to stroke severity, early neurological deterioration, and functional recovery, although the relationship is bidirectional and fraction‐specific, likely reflecting bilirubin's dual antioxidant and potentially neurotoxic properties at different concentrations [14, 15, 16, 17]. Renal function markers that did not differ significantly between groups in our bivariate analysis, such as creatinine and derived eGFR, have nonetheless been consistently shown elsewhere to independently predict in‐hospital and long‐term mortality, disability and recurrent vascular events after stroke, particularly when tracked longitudinally rather than at a single time point [20]. Likewise, abnormal liver enzymes (AST, ALT) in the acute and subacute phases have been associated with stroke severity and complications such as postprocedural hepatic dysfunction after thrombectomy, and, in some cohorts, paradoxically with smaller infarct volumes, suggesting a complex, time‐dependent hepato‐cerebral crosstalk [6, 21]. Taken together, this evidence supports the biological and clinical plausibility of using routine, low‐cost hematological and hepatorenal indices, particularly RDW‐CV, HCT, and bilirubin fractions, as adjunctive, accessible prognostic markers for risk‐stratifying stroke survivors in resource‐limited settings such as ours, where advanced neuroimaging‐based prognostic tools may not be readily available.

The alterations in RBC indices reflect a complex picture of erythrocyte remodeling. Lower MCHC can indicate iron deficiency, while high MCV is often linked to B₁₂/folate deficiency or alcohol use [22, 23]. The adjustment for confounders, particularly occupation (which may influence nutritional status), reversed the MCHC association and attenuated the MCV difference. This suggests that the observed bivariate differences may have been heavily influenced by prestroke nutritional or lifestyle factors that were correlated with occupation, rather than by the stroke itself. The final adjusted model showing higher MCHC with lower MCV and HCT could hint at a population of smaller, denser RBCs in stroke survivors. This phenotype is sometimes associated with cardiovascular risk and impaired deformability, potentially affecting cerebral microcirculation [24].

This study contributes significant baseline information regarding stroke to the population‐specific Ghanaian community, where such data are scarce. The rigorous laboratory methodology used, the adjustment for key confounders and adherence to STROBE guidelines make the study findings more reliable. However, the single‐center, hospital‐based recruitment may introduce selection bias. In addition, the case–control design precludes establishing causality. The exclusion of residual confounding factors like diet, specific medications, subclinical infections, or the exact stroke subtype and severity is another limiting factor. In light of the modest sample size, the findings should be regarded as preliminary and hypothesis‐generating, warranting confirmation in larger, adequately powered, multicenter studies before being applied clinically. The absence of subtype‐, severity‐ and treatment‐stratified analysis is an important limitation since these factors are known to differentially influence hepatorenal and hematological profiles.

5. Conclusion

In conclusion, this study suggests that stroke survivors in this Ghanaian population exhibit distinct hematological profiles, characterized by alterations in RBC morphology and possibly low‐grade hemolysis or altered bilirubin metabolism, independent of major liver or renal dysfunction. Clinically, these findings suggest that routine, low‐cost full blood count and bilirubin measurements, already widely available even in resource‐limited settings, may offer additional, easily accessible information for risk‐stratifying stroke survivors during rehabilitation, complementing clinical assessment. The profound influence of confounders such as occupation on these associations further underscores the critical role of socioeconomic and lifestyle factors in the stroke phenotype in this setting, and suggests that nutritional and occupational rehabilitation support may be as important as pharmacological management in this population. Future research should prioritize larger, multicenter, prospective studies that stratify by stroke subtype and severity, incorporate serial rather than single time‐point measurements of hepatorenal and hematological markers, and directly correlate these markers with validated functional outcome measures (e.g., modified Rankin Scale, NIHSS) and longer‐term survival, to establish their independent prognostic value and clinical utility in sub‐Saharan African stroke populations.

What is already known on this topic

• Stroke remains one of the leading causes of mortality and long‐term disability among adults globally.

• Demographic transitions, migration, and changes in dietary habits are causing a rise in the prevalence and incidence of stroke in low‐ and middle‐income countries.

• Stroke is not merely a cerebrovascular disease, but may also involve other organ systems, including the hepatic and renal systems.

What this study adds

• The study provides data on the hepatic and renal involvement of stroke regarding an indigenous African population, where such data are rare.

This study suggests that stroke survivors may exhibit distinct hematological profiles, characterized by alterations in RBC morphology and possibly low‐grade hemolysis or altered bilirubin metabolism independent of major liver or renal dysfunction.

Author Contributions

M.B., L.A.A., F.E., A.A., B.K.N., S.S.H., E.L.N., A.D.B., P.B.B., B.A., Y.A.K., P.D., T.A.: conceptualization and methodology; M.B.: supervision, administration, and validation; M.B., L.A.A., F.E., A.A., B.K.N., S.S.H., E.L.N., A.D.B., P.B.B., B.A., Y.A.K., P.D., T.A.: experimentation, data collection, and data curation; M.B., N.‐A.Y., J.A., M.M.M., J.A.A., A.‐W.D., E.O.A.: statistical analysis and writing‐draft. All the authors reviewed the draft manuscript and approved its content.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: STROBE Statement—Checklist of items that should be included in reports of case–control studies.

Data Availability Statement

The data supporting the findings of this study can be obtained from the corresponding author upon requests.