Overview of the study
A total of 315 febrile patients with confirmed P. falciparum malaria were followed for two seasons during high malaria transmission. The study excluded 66 individuals who were not within the catchment area, did not match the inclusion criteria, had severe G6PDd, or declined to participate (Figure 1). The remaining 249 (79%) patients were eligible and completed the 28-day follow-up. The initial characteristics of the two treatment groups were comparable (Table 1). Patient sociodemographic characteristics included the total number of patients who received ACT alone (n = 83) or ACT+SLD-PQ (n = 166) for uncomplicated P. falciparum malaria infection. Adults (aged > 15 years) accounted for the majority of the study population, with a total of 197 (79.12%) individuals and a sex ratio of males to females of 1.8 (Table 1).
Table 1: Baseline sociodemographic, clinical,parasitological profiles and prognostic profiles of the study participants (n = 249) by treatment group in Arjo-Didessa, southern Ethiopia; 2019-2022
Characteristics
|
Treatment arms
|
P- value
|
ACT alone
83 (33.33%)
|
ACT+SLD-PQ
166 (66.67%)
|
Gender (% male no. of males/total no. of individuals)
|
65.06(54/83)
|
64.46(107/166)
|
0.9257
|
Age(years), median (IQR)
|
20(13)
|
20 (15)
|
0.9276
|
Age groups
<15 years, no. (%)
>15 years, no. (%)
|
14(16.87)
69(83.13)
|
37(22.29)
129(77.71)
|
0.405
|
Axillary temperature (oC), mean (SD)
|
38.13(0.54)
|
37.98(.63)
|
0.0759
|
Fever (≥37.5oc) at present, no. (%)
|
63(75.90)
|
119(71.69)
|
|
Asexual parasite density/microliter, mean (SD)
|
13815.67(19655.76)
|
17430.92(30420.42)
|
0.3259
|
G6PD activity**, no. (%)
-Normal
-Deficient
|
77(92.77)
6(7.23)
|
155(93.37)
11(6.63)
|
|
Hb concentration at enrollment (g/dl), mean (SD)
|
12.80(1.18)
|
12.75(0.89)
|
0.736
|
G6PDd Hb(g/dl), mean (SD)
G6PDn Hb(g/dl), mean (SD)
|
12.70(0.88)
12.80(1.21)
|
12.7(0.53)
12.76(.91)
|
1.0000
0.730
|
Abbreviations: IQR, interquartile range; °C, degree Celsius, number [Note: means and standard deviation are presented for temperature and hemoglobin; age is presented as median and interquartile range]. ** as determined by a POCT analyzer indicating G6PD enzyme activity (IU/g Hb)].
Phenotypic G6PD status
At the time of enrollment, phenotypic point-of-care testing was utilized to screen patients for G6PD deficiency. The adjusted median G6PD enzyme activity was determined for the male participants by excluding 5 patients whose enzyme activity was less than 10% (less than 0.627 U/g Hb) of the median value found for all male participants. The median G6PD activity was 6.32 U/g Hb for all study participants (the range was 0.4–15.22 U/g Hb). The AMM (range) enzyme activity was 6.32 IU/gHb (0.67–15.219), which was 100%. The distribution of enzymatic activity by sex is illustrated in Figure 2. There was a distinct bimodal distribution in males and a unimodal distribution in females. When the participants' enzyme activity was between 0.4 and 1.896 U/g Hb (less than 30% of the adjusted male median), they were classified as deficient. Phenotypic G6PD data were recorded for 249 patients; 17 (6.83%) had G6PDd—14 males and 3 females. In the ACT alone arm, there were 6 (2.41%) patients with phenotypic G6PDd, whereas there were 11 (4.42%) patients in the ACT + SLD-PQ arm (Table 1).
Hemoglobin profiling
Data from 415 Hb measurements in 83 patients treated with ACT alone and 830 Hb measurements in 166 patients treated with ACT+SLD-PQ were used for the assessment of the changes in the mean Hb level between baseline and days 7, 14, 21, and 28. During enrollment, the mean (SD) Hb level was 12.80 (1.18) g/dL for patients receiving ACT alone and 12.75 (0.89) g/dL for those receiving ACT+SLD-PQ (Table 1). There was no significant difference in the baseline Hb levels between the two treatment arms (p = 0.7357). However, males had greater mean (SD) Hb concentrations at baseline (12.78 (1.03) [95% CI 12.62, 12.94]) than females did (12.75 (0.93) [95% CI 12.55-12.94]). There were no significant differences in the mean Hb concentration between males and females in the two treatment arms (p = 0.8281). However, as illustrated in Figure 3, in both groups receiving ACT+SLD-PQ, the mean Hb concentrations, shown as the mean change, decreased in the first week following treatment and reverted to baseline values during follow-up. Similarly, paired analysis of Hb concentrations relative to baseline demonstrated a reduction in the first week after receiving 0.25 mg/kg PQ + ACT or ACT alone. These reductions were significant on day 7 following SLD-PQ (0.25 mg/kg PQ) (P < 0.01) or ACT alone (P < 0.01; Table 2). However, there was no significant difference between the treatment groups, ACT alone and ACT+SLD-PQ (P = 0.157; Table 3).
Table 2. Mean change and mean hemoglobin concentration at baseline and day 7 in the treatment group in Arjo-Didessa, southern Ethiopia; 2019-2022
Treatment groups
|
N
|
Mean (Baseline)
|
Mean (on day 7)
|
Mean Diff
|
t -value
|
P value
|
ACT alone
ACT+ SLD-PQ
|
83
166
|
12.79639
12.7512
|
12.49277
12.29699
|
0.3036145
0.454216
|
3.6
13.75
|
0.0005*
0.0000*
|
Key: ACT, artemisinin combination therapy; PQ, primaquine. [Note: n, number of observations; mean baseline, mean of baseline hemoglobin concentration in g/dl; mean on day 7, mean of day 7 follow-up time hemoglobin concentration in g/dl; Diff; mean difference (95% CI) estimated using a paired Student t test comparing the mean concentration of hemoglobin on day 7 in each group].
After posttreatment (D7), Hb levels decreased on average by 0.45 g/dl (95% CI =0.39 to 0.52) in patients receiving ACT +SLD-PQ compared to 0.30 g/dl (95% CI =0.14 to 0.47) in patients receiving ACT alone (P = 0.157; Table 3). Patients with and without SLD-PQ experienced overall (D0–14) Hb losses of 0.10 g/dl (95% CI = -0.00 to 0.20) and 0.05 g/dl (95% CI = -0.123 to 0.22), respectively (P = 0.412; Table 3). At the overall follow-up (D0-28), the mean Hb levels were marginally lower in patients treated with ACT+SLD-PQ than in those not treated with ACT+SLD-PQ (Figure 3). Patients in the ACT alone and ACT + SLD-PQ arms recovered their Hb to or above the baseline values by day 28 (Figure 3).
Mean Hb (SD) changes following treatment
The mean reductions in Hb were [-0.05 ( 0.79)] g/dl and [-0.10 (0.68)] g/dl in the ACT alone and ACT + SLD-PQ arms, respectively, on day 14. The mean reductions in Hb were 0.04 (0.90) g/dl and 0.04 (0.61) g/dl for the ACT alone and ACT + SLD-PQ arms, respectively, on day 21. The day 28 mean reductions in Hb were 0.12 (0.95) g/dl and 0.00 (0.55) g/dl for the ACT alone and ACT+SLD-PQ arms, respectively. The mean Hb concentration gradually recovered and was close to baseline on day 28. The Hb level in the ACT alone group recovered by 0.25 (0.81) g/dl and 0.35 (0.66) g/dl for the ACT+SLD-PQ arm at the same follow-up day. The Hb level in the ACT alone group recovered by 0.35 (0.81) on day 21 and by 0.41 (0.63) g/dl on the same follow-up day for the ACT+SLD-PQ arm. The Hb recovery rates were 0.42 (0.86) and 0.45 (0.59) g/dl on day 28 in the ACT alone and ACT+SLD-PQ arms, respectively. The overall pattern showed an initial rapid drop in Hb levels followed by a lengthy recovery period; however, the difference between the two arms was not statistically significant (p = 0.157) (Table 3). On day 7, the mean Hb reduction was comparable between the groups (Table 3); for ACT alone and ACT+SLD-PQ, the difference (ACT + SLD-PQ-ACT alone) was 0.20 g/dL (95% CI -0.08–0.47; P = 0.157) after adjusting for baseline.
Table 3. The mean Hb concentration in each treatment arm and estimation of the mean Hb change and differences from the baseline at each follow-up period using the mixed model in Arjo-Didessa, southern Ethiopia; 2019-2022
|
Mean Hb Concentration (SD)
|
Mean Hb change (SD)
|
mean Hb change difference(95% CI)
|
Day
|
ACT alone
|
ACT+SLD-PQ
|
ACT alone
|
ACT+SLD-PQ
|
(ACT+SLD-PQ)- (ACT alone)
|
P value
|
0
|
12.80(1.18)
|
12.75(0.89)
|
*
|
*
|
*
|
*
|
7
|
12.49(1.16)
|
12.30(0.95)
|
0.30(0.77)
|
0.45(0.43)
|
0.20(-0.08, 0.47)
|
0.157
|
14
|
12.75(1.01)
|
12.65(0.78)
|
0.05(0.79)
|
0.10(0.68)
|
0.10(-0.13, 0.33)
|
0.412
|
21
|
12.84(.97)
|
12.71(0.82)
|
-0.04(0.90)
|
0.04(0.61)
|
0.13(-0.10, 0.36)
|
0.270
|
28
|
12.92(.93)
|
12.75(0.85)
|
-0.12(0.95)
|
0.00(0.55)
|
0.17(-0.07, 0.40)
|
0.161
|
|
Key: ACT, artemisinin combination therapy; 95% CI, confidence interval; Hb, hemoglobin; SLD-PQ, single low-dose primaquine. The mean absolute change was defined as the mean Hb level on the day of follow-up minus the mean Hb level on day 0. * baseline mean Hb only.
Hb reduction by G6PD phenotype and treatment arm
To address concerns about hemolysis associated with SLD-PQ use in G6PDd individuals, Hb concentrations were assessed at enrollment and throughout follow-up. Table 4 shows that the mean Hb concentration decreased with the G6PD phenotype. When all the G6PDd patients were phenotypically combined, compared with those in G6PDn patients, the mean Hb reduction was -0.24g/dl (P = 0.359). Overall, the baseline mean (±SD) Hb concentrations were similar between G6PDd patients and G6PDn patients [12.77 (1.01) g/dL vs. 12.70 (0.64) g/dL] and during each follow-up period [D7 12.14 (0.66) g/dL vs. 12.38 (1.05) g/dL; D14 12.45 (0.75) g/dL vs. 12.70 (0.87) g/dL; D21 12.54 (0.78) g/dl vs. 12.77 (0.88) g/dL; D28 12.62 (0.79) g/dl].
Mean Hb concentration reduction by G6PD phenotype
A paired t test model showed a negative association between D7-D0, D14-D0, D21-D0, and D28-D0 decreases in the G6PDd and G6PDn treatment groups, but G6PDn, in comparison to G6PDd, was linked to a positive change in D28-D0 Hb, with a mean increase of 0.05 g/dl (95% CI: -0.14, P = 0.316). The decrease in the mean Hb concentration after D 7 was greater in the G6PDd group than in the G6PDn group, but the difference was not significant (P =0.359): -0.56 g/dl vs. -0.39 g/dL, respectively; ΔHb concentration = -0.24 (-0.75, 0.27) g/dL. Table 4 shows the reduction in the mean Hb concentration according to the G6PD phenotype. On day 14, the mean decreases in Hb in the G6PDn and G6PDd groups were nearly equal to their values from the previous day, which were -0.07 g/dl and -0.25 g/dl, respectively. On day 21, however, there was no mean Hb decrease in G6PDn individuals, but there was a mean Hb decrease of -0.16 g/dl in G6PDd individuals. On day 28, the mean Hb reduction in G6PDd patients was -0.102 g/dl; however, the mean increase in Hb in G6PDn patients was 0.05 g/dl. This difference was not significant (P= 0.337).
Table 4. The mean hemoglobin concentration and range of change from baseline for patients with and without G6PD deficiency on day 7 after treatment.
Group
|
No. participants
|
Mean Hb
|
P value
|
G6PDd
|
17
|
12.141
|
|
G6PDn
|
232
|
12.379
|
0.359
|
Keywords: G6PDdd, glucose-6-phosphate dehydrogenase deficiency; G6PDn, normal glucose-6-phosphate dehydrogenase.
The mean changes in Hb levels on day 7 were slightly greater in G6PDd (from 12.70 g/dl to 12.14 g/dl; Δ mean = 0.56 g/dl) individuals than in G6PDn (from 12.77 g/dl to 12.38 g/dl; Δ mean = 0.39g/dl) individuals. were also observed.
In comparison to those with G6PDn ACT alone, both the ACT alone and the ACT+ SLD-PQ G6PDd cohorts experienced a lower mean Hb concentration, with mean changes ranging from 0.45 g/dl [95% CI: -0.486 to 0.079] to 0.54 g/dl [95% CI: -1.067 to 0.238] (Table 5 and Figure 4). In addition, on day 7 following ACT+SLD-PQ treatment, the Hb concentration in these G6PDd participants ranged from 12.70 to 12.10 g/dL, and that in the ACT-alone G6PDd participants ranged from 12.70 to 12.22. The distribution of patients between treatment arms was unaffected by sex or phenotypic G6PD phenotype. Although the mean changes in Hb in these groups were greater than those in the G6PDn ACT treatment group, a significant difference in Hb levels was not detected on day 7 posttreatment for G6PDd ACT+SLD-PQ vs. G6PDn ACT alone (P = 0.109, P = 0.304; Table 5).
Table 5. Results of multiple linear regression showing the mean Hb concentration and range of change from baseline for patients with and without G6PD deficiency on day 7.
Day 7
|
Coef.
|
t value
|
[95% CI]
|
p- value
|
Sex: base male
Female
Age
G6PD status
G6PDd ACT alone
G6PDn ACT+SLD-PQ
G6PDd ACT+SLD-PQ
|
0
-0.23
0.01
-0.45
-0.20
-0.54
|
.
-1.64
1.71
-1.03
-1.43
-1.61
|
.
-0.50-0.05
-0.00-0.02
-1.31-0.41
-0.48-0.08
-1.19-0.12
|
.
0.102
0.089
0.304
0.154
0.109
|
Keys: G6PD-d, glucose-6-phosphate dehydrogenase deficiency; G6PD-n, normal glucose-6-phosphate dehydrogenase. [Note: Coef is the mean Hb coefficient of variation estimated using a multiple linear regression model comparing the mean reduction in Hb concentration on day 7 in each group.]
However, the absolute mean reduction in Hb levels on day 7 was not significantly lower in either G6PDd (-0.54 g/dl 95% CI: -1.19, 0.12; P = 0.109) or G6PDn (-0.20 g/dl 95% CI: -0.48, 0.08; P = 0.154) ACT+SLD-PQ individuals (Table 5). There were consistently lower Hb concentrations in G6PDd participants treated with ACT+SLD-PQ than in G6PDn participants, although these differences were not clinically significant.
In the weeks following treatment with ACT alone or ACT + SLD-PQ, the Hb concentration relative to the total concentration gradually decreased; however, it initially decreased in both treatment groups and remained lower until day 28 (Figure 4). On day 14, the mean change in Hb concentration among G6PDd individuals receiving ACT+SLD-PQ was -0.45 (95% CI, -1.01 - 0.10, P= 0.108), whereas their G6PDn counterparts experienced only a minor change of -0.10 (95% CI, -0.34 -0.13; Table 6). There was no significant difference in the mean Hb concentration between G6PDd ACT-alone patients and G6PDn ACT-alone patients (P = 0.234).
Table 6. The mean hemoglobin concentration and range of change from baseline to day 14 for patients with and without G6PD deficiency.
Day 14
|
Coef.
|
t value
|
[95% CI]
|
p value
|
Sex: base male
Female
Age
G6PD status
G6PDd ACT alone
G6PDn ACT+SLD-PQ
G6PDd ACT+SLD-PQ
|
0
-0.016
0.012
-0.44
-0.10
-0.45
|
.
-1.40
2.29
-1.19
-0.85
-1.61
|
.
-0.39 -0.07
0.00 - 0.02
-1.17- 0.29
-0.34- 0.13
-1.01 - 0.10
|
.
0.163
0.023
0.234
0.397
0.108
|
Keywords: G6PD-d, glucose-6-phosphate dehydrogenase deficiency; G6PD-n, normal glucose-6-phosphate dehydrogenase. [Note: Coef, mean Hb coefficient of variation estimated using a multiple linear regression model comparing the mean reduction in Hb concentration on day 14 in each group.]
In this study, we first reported changes in the mean Hb concentration relative to the baseline value (Figure 3) and the mean changes in the Hb concentration relative to the G6PD status (Figure 4). Mean hemoglobin concentrations, expressed as an absolute or relative change, decreased in the first week following therapy in all ACT+SLD-PQ-treated groups and recovered to baseline levels at follow-up (Figure 4). Similarly, one week after the administration of ACT+SLD-PQ, G6PDd individuals showed a decrease in their hemoglobin concentration compared to that at baseline (Figure 4). The mean Hb levels did not significantly differ between the patients treated with ACT alone and those treated with ACT+SLD-PQ on day 7 (coefficient, -0.45; 95% CI, -1.01, 0.10; P = 0.108).
Severe adverse events in the treatment groups
In our study, abdominal pain, appetite loss, fatigue, and nausea were the major AEs, followed by skin rash, cough, headache, diarrhea, and vomiting. Out of the 249 study participants who completed the follow-up, 94 AEs were recorded—51 (30.7%) from the ACT+SLD-PQ cohort and 43 (51.8%) from the ACT alone group (Table 5). All of the reported AEs were rated as mild. The difference in the incidence of adverse events (AEs) between the treatment groups was not significant (Table 7). There were no SAEs in either treatment group. None of the participants required blood transfusions. In addition, during the study follow-up, every adverse event (AE) resolved, and none of the adverse events stopped participating.
Table 7. Adverse Events (AEs) Among Study Participants by G6PD Status and Treatment Group, Arjo Didessa, Ethiopia
|
Treatment arms
|
Adverse events
|
ACT only
|
ACT+SLD-PQ
|
p- value
|
G6PDn
n (%)
|
G6PDd
n (%)
|
G6PDn
n (%)
|
G6PDd
n (%)
|
|
Head ache
|
4(11.43%)
|
0
|
1(2.38%)
|
1(11.11%)
|
0.097
|
Nausea
|
6(17.14%)
|
1(12.5%)
|
4(9.52%)
|
1(11.11%)
|
0.062
|
Vomiting
|
1(2.86%)
|
0
|
1(2.38%)
|
1(11.11%)
|
0.705
|
Abdominal pain
|
5(17.14%)
|
2(25%)
|
12(28.57%)
|
2(22.22%)
|
0.460
|
loss of appetite
|
5(14.29%)
|
2(25%)
|
9(21.43%)
|
1(11.11%)
|
0.321
|
Fatigue
|
6(17.14%)
|
1(12.5%)
|
5(11.91%)
|
1(11.11%)
|
0.098
|
Skin rashes
|
2(5.71%)
|
1(12.5%)
|
5(11.91%)
|
1(11.11%)
|
0.626
|
Cough
|
4(11.43)
|
1(12.5%)
|
3(7.14%)
|
0
|
0.084
|
Diarrhea
|
1(2.86%)
|
0
|
2(4.76%)
|
1(11.11%)
|
0.593
|
Total
|
35(100%)
|
8(100%)
|
42(100%)
|
9(100%)
|
|