Retinopathy of vivax malaria in adults and its relation with severity parameters

Abstract

Malarial retinopathy is a set of retinal signs in severe malaria due to falciparum malaria. With increased recognition of severe manifestations of vivax malaria, a systematic study to evaluate retinal changes in vivax malaria could elaborate our knowledge about this neglected entity. This observational study included retinal examination of 104 adult patients (>14 years) with varying severity of vivax malaria admitted to a tertiary care center during peak seasons of 2012 and 2013. Thirty-eight percent of severe cases had a retinal sign as compared to 6% of non-severe cases (p < 0.01). No statistically significant effect of residence or age on the presence of retinopathy was noted. Females were found to be more prone to develop a retinal sign (p < 0.01). Presence of retinal signs was significantly associated with anemia and jaundice. No statistical association was noted for retinal signs to be present in either renal dysfunction or altered thrombocytes count. The most common signs were arteriovenous changes, present in eight cases (19%) of severe malaria and three cases (5%) of non-severe malaria. Retinal hemorrhage was present in five cases (12%) of severe malaria and no case of non-severe malaria. Both superficial and deep hemorrhages were seen including white-centered hemorrhages. Other signs included cotton wool spots, hard exudates, blurred disk margins with spontaneous venous pulsations and bilateral disk edema. A correlation between retinal signs and severity parameters was drawn from the study. This is the first systemic study to evaluate the retinal changes in vivax malaria. Larger prospective studies should be done for further knowledge regarding retinal changes in vivax malaria, especially severe disease. Apart from its clinical significance, it might lead to a better understanding of the pathogenesis of the systemic disease of vivax malaria.

Introduction

Malaria is the most life-threatening parasitic disease of human beings. Present globally, an estimated 3.4 billion people living in 104 countries were at risk in 2013.¹ Among the five species of Plasmodium genus causing malaria, P. falciparum and P. vivax are the most important. Although malaria due to P. falciparum is the most deadly form, worldwide P. vivax has a wider distribution than P. falciparum.¹ An estimated 2.48 billion people lived at a risk of P. vivax in 2010.² In spite of disagreement regarding global incidence of clinical cases of P. vivax, with some estimates as high as 132–391 million cases per year, its enormous burden on global health is being increasingly appreciated.³ The overall global cost of P. vivax infection has been conservatively estimated to be between US $ 1.4 and 4.0 billion per year.³

India is a major contributor to the global burden of vivax malaria with about 50% of global population at risk residing here.² Due to low and erratic, unstable transmission in India, the affected population achieves little immunity to this parasite and consequently, people of all ages are at risk here. Although national figures show almost equal incidence of the two species, individual areas differ considerably in proportion of cases due to the two species. P. vivax contributes to more than 90% cases of malaria in the northwestern state of Rajasthan.⁴ P. vivax population in India is highly diverse in terms of phenotypic traits such as relapse patterns, drug response, and clinical profiles and is highly genetically variable according to studies of antigenic genes, isozyme markers, and microsatellites markers.⁵ The presence of harsher weather and terrain in certain areas favor the presence of more robust species and provide unique incentives for studying P. vivax in India.⁵

Various reports in last decade, including those from this part of India, regarding morbidity and mortality related to vivax malaria have challenged the long-held dogma of it being a ‘benign’ disease^6,7 and have lead to the recognition of severe and complicated manifestation vivax malaria including cerebral malaria. The paradigm of P. vivax causing benign disease has been challenged with evidence suggesting that P. vivax itself can contribute significantly to mortality. The neglect of this disease is being increasingly recognized. The role of inflammatory imbalance in the pathogenesis of severe disease is being considered. Even sequestration of red blood cells (RBCs) infected with P. vivax has been shown in few experimental studies. Possibility of sequestration can also be calculated indirectly with other observation like low birth weight in children born to women suffering from vivax malaria during pregnancy.⁸ It has been seen that low parasite densities are sufficient to produce severe disease in P. vivax malaria. Moreover, P. vivax can persist as dormant hypnozoites in liver, which bedsides causing clinical relapses, enables parasite to survive adverse climate and maintain reservoir. This factor, along with current unavailability of any diagnostic method for detecting them and inadequacy of the only available therapy against relapse, limits the feasibility of regional elimination of P. vivax.⁹ Thus, increased the need of characterization of the disease related to P. vivax, especially severe disease in different geographical areas has long been felt.³

Malarial retinopathy is a set of retinal signs which includes retinal whitening, vessel changes in the form of discoloration to orange or white, retinal hemorrhage, and papilloedema, which have been well characterized in African children suffering from cerebral malaria caused by P. falciparum.¹⁰ Detection of malarial retinopathy in a comatose child in Africa is now considered as important diagnostic criteria of cerebral malaria. In an autopsy study, ophthalmologists demonstrated a collective sensitivity of 95% and specificity of 90% when using the detection of retinopathy as a diagnostic test for cerebral malaria in subsequently fatal cases of coma.¹¹ This can be compared with a specificity of 61% for the diagnosis of cerebral malaria using all other available clinical and laboratory data. Patients with parasitemia, but no malarial retinopathy, were found at postmortem to have evidence of alternative causes of death and no significant cerebral sequestration. Considering only parasitemic patients, detection of any malarial retinopathy improved specificity of the cerebral malaria diagnosis from 61 to 100%, hence, the importance of funduscopy examination in cases of malaria. In addition, retinal changes in adults with severe falciparum malaria (both cerebral and non-cerebral) have also been described from this region.^12–14

There have been few isolated case reports regarding eye changes in cases of vivax malaria including retinal hemorrhages.^15–18 The presence of retinal changes assumes the importance in light of increased recognition of severe disease in vivax malaria.

Considering the significance of P. vivax malaria in this region and the recognition of severe manifestations of the disease, the current study was undertaken to evaluate the spectrum of retinal manifestations in cases of P. vivax malaria and to find out the relation of retinopathy with various biochemical and hematological indices in adults.

Patients and methods

Bikaner is situated in the northwestern part of Rajasthan, India between longitude 71°54′ and 74°12′ east and latitude 27°11′ and 29°3′ north. Its altitude is 237 m above the sea level. The region experiences a hot summer, a cold winter, and a low annual rainfall of 25 cm. Most of the rains occur in a single period between July and September. This is basically an arid zone, which had experienced changes in ecosystem in the last quarter of century which has lead to alteration in malaria transmission. Malaria cases are seen throughout the year with marked seasonal variation. Maximum cases occur in the post-rainy season and least number of cases occurs in January and February.¹⁹

The study was approved by the institutional review board. Study sample comprised of adults (>14 years of age) both male and female, admitted to various medicine wards of PBM Hospital, Bikaner – a tertiary care institution from 2012 to 2013 with a diagnosis of malaria due to P. vivax monoinfection established by peripheral blood film examination for the parasite and/or rapid diagnostic tests and confirmation by PCR for presence of P. vivax and absence of P. falciparum.

Informed consent was taken from patients or attending relatives in case of minor patients (14–18 years). After detailed history and physical examination, all patients were subjected to thorough ophthalmological examination during the stay at hospital in a single sitting which included direct and indirect ophthalmoscopy after pupil dilatation with tropicamide 0.8% and phenyl ephrine 5% eye drops. Fundus photography was done in selected cases after stabilization of condition or at discharge.

Blood samples were collected for hematological and biochemical investigations hemoglobin, hematocrit, RBC count, white cell count, platelet count, thick and thin peripheral blood smears, ESR, glucose levels, blood urea, serum creatinine, serum bilirubin, and liver enzymes. Bleeding time, clotting time, and prothrombin time were done in patients with bleeding tendency. Widal test, Leptospira, HBsAg and HCV, dengue serology, HIV, serum electrolytes, lumbar puncture for CSF, ECG, chest X ray, abdominal USG, CT head, MRI were done in appropriate cases.

The patients were managed according to the WHO protocol and national protocol. Adequate antiobiotic therapy and supportive care including hemodialysis to patients of acute kidney injury was given.

Patients not willing to give consent, unable or unwilling to cooperate with eye examination, contra indicated to tropicamide eye drops (angle closure glaucoma or documented allergy), having severe corneal scarring or cataracts in both eyes, diabetes mellitus, hypertension, meningitis, encephalitis, epilepsy, head injury, intra-cranial space occupying lesion, chronic renal disease, chronic liver disease, preexisting coagulopathy, etc. were excluded from the study after systematic evaluation.

The data from patients satisfying the inclusion criteria were tabulated and analyzed using chi-square test.

Results

Study sample

One hundred and fifty cases with varying severity of malaria were examined for fundus changes during study period, out of which 46 were excluded since they did not fulfill eligibility criteria given in method section. Age range of the study sample was 15–65 years (mean ± SD = 30 ± 12 years) with 59% patients being male and 62% of patients were less than 30 years of age. There was no statistical difference between male and female groups with respect to age.

Severe disease²⁰ was present in 42 cases (17 male and 25 female) (Table 1). Females were present in much higher proportion in severe group (60%) than non-severe group (31%) and were found to be more prone to develop severe disease compared to males (p value < 0.01).

Table 1.

Distribution of severity in patients with malaria (n = 104)

Severe/complicated (n = 42)	No (%)
Anemia (Hb ≤ 7 g/dl or PCV ≤ 20)	18a (43)
Renal dysfunction (S. Creatinine ≥ 3 mg/dl)	5 (12)
Jaundice (S. Bilirubin ≥ 3 mg/dl)	10 (24)
Shock (Systolic BP < 80 mm Hg)	1 (2)
Thrombocytopenia (<50 × 1000/cu mm)	26 (62)
Septicemia	1 (2)
Moderate (n = 49)	No
Anemia (Hb = 7–10 g/dl or PCV = 20–30)	19
S. creatinine = 1.2–3 mg/dl	8
S. bilirubin = 1.2–3 mg/dl	3
Thrombocytopenia (50 K to 1 L)	41
Uncomplicated malariab (n = 13)	–

^aIncluding one patient with microcytic hypochromic anemia.

^bIncluding mild anemia (Hb ≥ 10 g/dl or PCV ≥ 30) and mild thrombocytopenia (100–150 × 1000/cu mm).

Retinal changes were present in 20 cases (19% of study sample), out of which 16 had severe disease and 4 had non-severe disease (Tables 2 and 3). Overall 38% of severe cases had a retinal sign as compared to 6% of non-severe cases (p < 0.01). Ten percent of all males had retinal signs in comparison to 33% of all females. Thus, females were more prone to develop a retinal sign (p < 0.01). This was probably because females in this study were more prone to develop severe disease. Distribution of retinal signs according to severity of malaria is shown in Table 4 and according to individual severity parameters is shown in Table 5.

Table 2.

Clinical profile of cases with retinopathy

								Ocular findings
								Retinal findings
S. No.	Case No.	Age (Yrs.)	Sex	Rural/urban	Specific feature	Remarks		Right eye	Left eye
1	1	24	F	Rural	Moderate anemia, mild thrombocytopenia	Moderate	Pallor, LE subconjunctival hemorrhage	NAD	Hard exudate = 1
2	9	25	F	Rural	Severe anemia	Severe	Pallor	Roth spots > 10	Roth spots > 10, subhyaloid hemorrhage over lying macula
3	10	55	M	Rural	Mild anemia, moderate thrombocytopenia	Moderate		Venous tortousity	Venous tortousity
4	13	22	F	Rural	Pregnancy 6th month, severe anemia	Severe	Pallor	Venous dilatation and tortousity	Venous dilatation and tortousity
5	15	28	M	Urban	Jaundice	Moderate	Icterus	NAD	Disk margin blurred, venous tortousity
6	18	25	F	Rural	Pregnancy 8th month, severe anemia	Severe	Pallor	Venous tortousity	Venous tortousity
7	24	47	F	Rural	Moderate anemia, moderate thrombocytopenia, jaundice	Severe	Icterus	Disk margin blurred on 2 sides (SVP +)	Disk margin blurred on 3 sides (SVP+)
8	26	25	M	Rural	Mod anemia, mild thrombocytopenia, septicemia	Severe	Pallor	Venous tortousity	Venous tortousity
9	28	40	M	Rural	Mild anemia, moderate thrombocytopenia, jaundice	Severe	Icterus	NAD	Hard exudate = 1
10	49a	42	M	Urban	Severe anemia, severe thrombocytopenia	Severe	Pallor	Retinal hemorrhage = 3	NAD
11	53b	15	F	Rural	Severe anemia, severe thrombocytopenia, severe AKI	Severe	Pallor	Disk edema, retinal edema, venous dilatation, arterial tortousity	Disk edema, retinal edema, venous dilatation, arterial tortousity
12	54	45	F	Rural	Severe anemia, mild thrombocytopenia, AKI, jaundice	Severe	Pallor, icterus	Hemorrhages = 10, cotton wool spot = 1, venous dilatation, arterial tortousity	Hemorrhages = 5–8, venous dilatation, arterial tortousity
13	63	31	M	Rural	Moderate Anemia, moderate thrombocytopenia, jaundice	Severe	Pallor, icterus	Venous dilatation	Venous dilatation
14	66	40	F	Rural	Moderate anemia, moderate thrombocytopenia	Moderate	Pallor	Venous tortousity, disk margin blured on 2 sides (SVP +)	Venous tortousity, disk margin blured on 2 sides (SVP +)
15	71	25	F	Rural	Severe anemia, moderate thrombocytopenia, jaundice	Severe	Pallor, icterus	2 white-centered hemorrhages, venous dilatation	2 hemorrhages, venous dilatation
16	72	25	F	Rural	Severe anemia, severe thrombocytopenia	Severe	Pallor	Disk pallor	Disk pallor
17	75	17	F	Urban	Severe anemia, moderate thrombocytopenia, jaundice	Severe	Pallor, icterus	Disk pallor	Disk pallor
18	79	30	F	Rural	Severe anemia, severe thrombocytopenia, severe AKI	Severe	Pallor	Hemorrhage = 1	NAD
19	81	25	F	Rural	Moderate anemia, severe thrombocytopenia	Severe	Pallor	Cotton wool spots = 2	NAD
20	82	21	F	Urban	Moderate anemia, severe thrombocytopenia	Severe	BE subconjunctival hemorrhage	Venous dilatation	Venous dilatation

^aCase reported⁹⁰.

^bPatient with microcytic hypochromic anemia.

Table 3.

Lab profile of cases with retinopathy

						Renal function test				Liver function test
						Blood urea (mg/dl)	Serum creatinine (mg/dl)	Serum billirubin		SGOT	SGPT	ALP
S. No.	Case No.	Hb (g/dl)	PCV (%)	TLC (per cu mm)	Platelet count (×1000/cu mm)			Total (mg/dl)	Direct (mg/dl)	(IU/L)	(IU/L)	(IU/L)
1	1	7.5	27	3500	120	26	0.9	0.9	0.2	24	25	110
2	9	3.4	10	5200	218	34	1.2	1.2	0.4	42	39	156
3	10	11	35	5000	58	32	1	1.1	0.2	45	30	110
4	13	7	24	6000	186	21	0.7	1	0.3	26	24	110
5	15	13.6	46	10200	225	22	0.6	2.2	1.3	175	190	250
6	18	7	27	5000	200	26	0.8	1.2	0.4	26	24	112
7	24	9.4	28	8300	50	26	0.7	7.2	4.5	NA	NA	NA
8	26	8.2	28	36000	103	30	0.9	0.7	0.2	29	32	156
9	28	11	35	5900	65	35	0.9	5.5	3.2	NA	NA	NA
10	49	5.6	16	8000	40	28	0.9	1.1	NA	40	35	111
11	53	6.1	16	7400	24	142	13.7	0.7	0.2	22	21	176
12	54	4.9	15	5400	130	148	2.3	3.7	NA	71	119	312
13	63	8	24	7400	51	42	1.2	3.7	1.8	63	177	85
14	66	7.8	26	3200	80	28	0.8	0.8	0.5	34	28	368
15	71	6.3	17	2250	75	35	1	11.2	6.2	31	44	241
16	72	3.6	15	2400	42	25	0.9	1.2	0.4	38	74	168
17	75	5.9	19	2100	55	28	0.8	4.8	3.8	124	111	503
18	79	4	14	6400	24	92	3.6	1.2	0.4	34	42	NA
19	81	7.9	28	3360	49	27	0.8	1.2	0.8	43	36	NA
20	82	9	31	4500	41	26	0.8	0.7	0.2	18	24	65

Table 4.

Distribution of retinal signs according to severity of malaria (n = 20)

Signs	Severe	Moderate	Uncomplicated	Total
No retinopathy	26	45	13	84
Any retinopathy	16	4	0	20
Any hemorrhage	5	0	0	5
White centered hemorrhage	2	0	0	2
Cotton wool spots	2	0	0	2
Hard exudates	1	1	0	2
Disk edema	1	0	0	1
Disk margin blurring	1	2	0	3
Disk pallor	2	0	0	2
Arteriovenous changes	8	3	0	11

All retinal signs were present in severe disease. Hemorrhages, cotton wool spots, disk edema, and disk pallor were present exclusively in severe disease. Only hard exudates, disk margin blurring, and arteriovenous changes (dilation and tortousity) were seen in non-severe disease.

Table 5.

Distribution of retinal signs according to various severity parameters

	Any	Retinal hemorrhage			Hard	Disk		Disk	AV changes (dilatation/tortousity) (n = 11)
Severity parameter	Retinopathy (n = 20)	Any (n = 5)	White centered (n = 2)	Cotton wool spots (n = 2)	Exudates (n = 2)	Edema (n = 1)	Blurred disk margin (n = 3)	Pallor (n = 2)
Anemia (Hb in g/dl)
Mild (≥10)	2	0	0	0	1	0	0	0	1
Moderate (7–10)	7	0	0	1	1	0	2	0	4
Severe (≤7)	10*	5	2	1	0	1*	0	2	5
Jaundice (S. Bilirubin mg/dl)
<3	1	0	0	0	0	0	1	0	1
≥3	6	2	1	1	1	0	1	1	3
Renal dysfunction (S. Creatinine mg/dl)
<3	1	1	0	1	0	0	0	0	1
≥3	2	1	0	0	0	1	0	0	1
Thrombocytopenia
Mild (1–1.5 L/cu mm)	3	1	1	1	1	0	1	0	2
Moderate (1–1.5 L/cu mm)	7	1	0	0	1	0	1	1	4
Severe (1–1.5 L/cu mm)	6	2	1	1	0	1	0	1	2
Septicemia	1	0	0	0	0	0	0	0	1
Pregnancy	2	0	0	0	0	0	0	0	2

^aOne patient had microcytic hypochromic anemia.

Severity indices and retinal changes

Severe anemia was present in 18 cases, out of which one patient had microcytic hypochromic anemia. She had bilateral disk edema with peripapillary edema, venous dilation, and arterial narrowing. Fifty-six percent of patients with severe anemia had a retinal sign as compared to 12% of non-severe anemia (p < 0.01) (Figure 1 and Table 6). Other retinal changes present in severe anemia cases were retinal hemorrhages in five cases including white-centered hemorrhages in two cases, cotton wool spots in one, disk pallor in two, and arteriovenous changes in five. Retinal signs in non-severe anemia included cotton wool spots in one, hard exudates in two and disk margin blurring in two and arteriovenous changes in five.

Figure 1.

Retinopathy: distribution according to hemoglobin levels (n = 104).

Table 6.

Retinopathy: distribution according to severty of anemia

	Any retinal sign			No retinal sign
Anemia	Male	Female	Total	Male	Female	Total	Grand total
No	1	0	1	9	0	9	10
Milda	2	0	2	36	4	40	42
Moderateb	2	5	7	10	17	27	34
Severec	1	9*	10*	0	8	8	18*
Total	6	14	20	55	29	84	104

Notes: 6% patients with Hb ≥ 10 g/dl had a retinal sign.

21% patients with Hb = 7–10 g/dl had a retinal sign.

56% patients with Hb ≤ 7 g/dl had a retinal sign.

85% patients with retinal sign had Hb < 10 g/dl.

Thus, the presence of retinal sign in severe anemia was statistically significant compared to non-severe anemia (p < 0.01).

Anemia: n = 94.

^aMild anemia: Hb = 10–12 g% in females, 10–13 g% in males.

^bModerate anemia: Hb = 7–10 g% or PCV = 20–30 in both male and female.

^cSevere anemia: Hb ≤ 7 g% or PCV ≤ 20 in both male and female.

^*One patient had microcytic hypochromic anemia.

Jaundice (S. bilirubin > 3 mg%) was present in 10 cases. The presence of retinal signs in cases with jaundice was found to be statistically significant (p value < 0.01). Jaundice was also associated with retinal hemorrhage in two including white-centered hemorrhage in one, cotton wool spots in one, hard exudates in one, and disk pallor and margin blurring in one each. Thus, presence of anemia and jaundice together seemed to have more risk of showing retinal signs. There was a young female patient having shock, moderate anemia, moderate thrombocytopenia, renal dysfunction, and jaundice also showed no sign on fundus examination. A young male patient having septicemia with moderate anemia and mild thrombocytopenia showed venous dilation and tortousity in both eyes. Venous dilation and tortousity in retinal vessels are evidence of sluggish blood flow in the veins (Figures 2–4).

Figure 2.

Fundus photorgraph showing bilateral disk edema, perpapillary edema, venous dilatation, and tortousity, arterial tortousity.

Figure 3.

Fundus photorgraph of right eye showing two cotton wool spots.

Figure 4.

Fundus photorgraph of right eye showing multiple retinal hemorrhage and venous dilatation.

No statistical association was noted for retinal signs to be present in either renal dysfunction or thrombocytopenia.

There were five females who were at different stages of pregnancy. Two had severe anemia alone and showed arteriovenous changes in form of dilation and tortousity (Table 5). Of the other three, first had moderate anemia, second had severe thrombocytopenia with moderate anemia, and third had severe anemia, severe thrombocytopenia, and severe renal dysfunction combined did not show any retinal sign. No definite correlation with the severity criteria could be made out in the pregnant females.

In summary, the most common observation was the evidence of arteriovenous changes in the form of dilatation and tortousity in eight cases (19%) of severe malaria and three cases (5%) of non-severe malaria. Second most common sign was retinal hemorrhage present in five cases (12%) of severe malaria and no case of non-severe malaria. Both superficial and deep hemorrhages were seen and two of the five patients had white-centered hemorrhages.

No patient had subhyaloid or vitreous hemorrhage. Cotton wool spots and hard exudates were present in two patients each. Blurred disk margins with spontaneous venous pulsations were present in three cases, whereas bilateral disk edema with venous dilatation and arterial attenuation was seen in one case of severe microcytic hypochromic anemia with renal failure (case 11). No patient had retinal whitening or vessel discoloration as seen in retinopathy due to falciparum malaria and no patient presented with any visual complaint except for one who had a retinal hemorrhage overlying fovea (case 10).

Discussion

With implementation of molecular diagnosis, it has become evident that P. vivax monoinfection could also be involved in multiple organ dysfunction and severe life-threatening disease as seen in P. falciparum infection.⁶ We believed that the study of retinal changes in vivax malaria could elaborate the knowledge regarding this neglected entity.

In our study, 38% of severe cases had a retinal sign as compared to 6% of non-severe cases (p < 0.01). Earlier in a similar study done in adult patients of falciparum malaria in the same region, Kochar et al. had found ophthalmoscopic findings to be present in 34% cases of cerebral malaria, 24% cases of non-cerebral severe malaria, and 12% cases of uncomplicated malaria.¹³ Another study on adults with severe falciparum malaria in Bangladesh (a low transmission area) found retinal changes to be present in 14 of 20 patients with cerebral malaria, 3 of 7 patients with non-cerebral severe malaria, and 3 of 15 patients with uncomplicated malaria.¹² In comparison studies, in children by Beare et al. and Essuman et al. found retinopathy in 53% children with non-cerebral severe malaria due to P. falciparum in Malawi and Ghana, respectively.^21,22

The presence of retinal signs was significantly associated with female sex, severe anemia, and jaundice. The presence of renal failure or thrombocytopenia was not associated with retinal changes. The most remarkable findings were presence of retinal hemorrhages in five cases including two with white-centered hemorrhages. There are only a few previous case reports regarding the same available on PubMed. Biswas et al. reported two cases of vivax malaria with subhyaloid hemorrhage from south India.¹⁵ Choi et al. and Lee et al. also reported one case each from South Korea while citing previous two other Korean cases.^16,17 The cases of vivax with retinal hemorrhages reported earlier had variable levels of hemoglobin and platelet counts. All five cases in our series had severe anemia.

Retinal hemorrhages have been typically ascribed to cerebral malaria caused by P. falciparum. Since the recognition that presence of retinal hemorrhages is associated with severity indicators, especially in cerebral malaria, it has been suggested that retinal hemorrhages may be visible evidence of vascular lesions involved in the pathogenesis of this condition.¹⁴

Significance of detection of retinal hemorrhages in cerebral malaria was shown by a histopathology study which correlated number of retinal hemorrhages to density of brain hemorrhages in children dying of cerebral malaria.²³

The mechanism of retinal hemorrhage is uncertain in P. vivax infection. Sequestration of parasitized RBCs and cytoadhesion with rosetting are key pathogenic processes in P. falciparum patients. Accompanied thrombocytopenia and anemia could play a role in the development of retinal hemorrhage. However, Incorvaia et al. have reported different available studies regarding influence of thrombocytopenia or lowered hematocrit per se on retinal hemorrhages and it was not conclusive.²⁴ Similarly, severe normocytic normochromic anemia has not been considered traditionally to be attributable to vivax disease, but its presence leading to retinal hemorrhage could indicate that the pathogenic processes underlying it may be benign.

The other remarkable finding was presence of cotton wool spots in two patients. Only 1 case of vivax malaria has been reported earlier to have cotton wool spot.¹⁷ The cotton wool spot, a disorder in axonal transport, is thought to develop because of the localized ischemia of retina.²⁵ This finding is important because cotton wool spot suggests a microcirculatory problem in the presumed absence of sequestration.

Arteriovenous dilation and tortousity were present in 19% cases of severe malaria and 5% cases of non-severe malaria, respectively. Earlier, Essuman et al also found these vessel changes in 22% cases of non-cerebral severe malaria due to P. falciparum in children in Ghana.²²

Tortuous retinal veins have been observed in a variety of systemic conditions affecting cellular and plasmatic composition of blood, altering flow characteristics, viscosity, or coagulability and have been implicated in conditions causing retinal hypoxia.²⁴ Essuman et al. suggested their presence in falciparum malaria to be possibly because of mechanical obstruction by parasites in either or both central and peripheral veins or due to some metabolic cause such as metabolic steal by intravascular parasites leading to hypoxic stress.²²

Hard exudates have chronic pathogenesis and seem to be incidental findings in our series.

Retinal whitening and vessel discoloration as seen in retinopathy due to falciparum malaria in children or both child and adults were not seen in our series. These signs have been hypothesized to be associated with sequestration of RBCs infected with P. falciparum causing mechanical obstruction.^26,27 Their absence in our study may probably be the result of the fact that sequestration is not a pathogenic mechanism in vivax malaria. However, the presence of some features of retinopathy similar to P. falciparum malaria, i.e. hemorrhages and Roth spots indicate that there are some common pathogenic mechanisms which are shared by the two parasites.

Venous dilation and tortousity and the presence of cotton wool spots probably indicate some hypoxic insult and these may occur in P. vivax malaria as well, probably by mechanisms other than mechanical obstruction such as mitochondrial failure in response to inflammatory mediators.²⁸

There is evidence of marked imbalance of inflammatory mediators in P. vivax malaria.^29,30 A recent report from Pakistan indicates upregulation of endothelial adhesion molecule ICAM along with TNFα in complicated vivax cases.³¹ These mediators have been hypothesized to be causing microvascular disruption in falciparum malaria.³² Possibly, these could be the common pathogenic mechanisms of the retinal changes among the two species. Moreover, cytoadhesion and rosetting of P. vivax-infected RBCs have also been demonstrated experimentally.^{33, 34} The contribution of these pathogenic mechanisms, albeit minor, cannot be ruled out completely.

To conclude, possibility of retinal changes could be expected in P. vivax malaria and the presence of retinal signs, especially hemorrhage in a febrile patient could suggest a diagnosis of P. vivax malaria to the treating physician.¹⁷ The retinal changes tend to be associated with severe manifestations and probably relate to the pathogenesis of severity or its effects. Further, the observations of retinal changes only in South Korea and India so far, could suggest clinical and parasitological differences in P. vivax between tropical and temperate areas.¹⁶

This is the first systemic study to evaluate the retinal changes in vivax malaria. Since this was a cross-sectional study, it was difficult to establish causality between P. vivax (and severe vivax) and retinal changes and the possibility of observed changes to be a result of deranged hematologic, renal, and hepatic functions cannot be ruled out. Therefore, larger prospective studies with longer follow-up should be done for the further advancement in our knowledge regarding retinal changes in vivax malaria, especially severe disease. Apart from its clinical significance, it might lead to a better understanding of the pathogenesis of the systemic disease of vivax malaria.