Automated erythrocytapheresis in the treatment of severe falciparum malaria

233

Case Reports In summary, the lessons to be learned include, antimalarial prophylaxis is essential in all travellers to malaria zones and the importance of considering Ihe diagnosis of malaria in a patient presenting with a febrile illness following appropriate foreign travel and finally reinforces the value of automated ET in seriously ill patients with PIN high parasitaemias and multiorgan complications. in conjunction with specific antimalarial and full supportive therapy.

References 1 Stone WJ, Hanchett JE. Knepshield JH. Acute renal insufficiency due to falciparum malaria: review of 42 cases. Arch b?tei-11 Med 1972: 129: 620-628. 2 Kurathong S. Srichailtul T. Isarangkura l? Phanichphant S. Exchange transfusion in cerebral malaria complicated by dissemnated intravascular coagulation. Soutilensl Asin~? J Troy? Mrd PrUc Hmlth 1979: 10: 389~392. 3 Phillips P. Mantel S. Renny WB. Exchange transfusion as an adjunct to the treatment of severe falciparum malaria: case report and review. Rev 111Jccr Dis 1990: 12: 1100-I 108. 4 Espinosa G. Tortajada C. Gascon J. Miquel R, Nicolas JM. Nadal P. Corachan 31. Severe plasmodiurn falciparum malaria. Description of 5 casts. RclJisln Cfill Esptrnofn 1997: 197: 631-634. 5 Lobe1 HO. Bernard KW Williams SL. Hightower AW Patchen 1X. Campbell CC. Effectiveness and tolerance of long term malaria prophylaxis with mefloquine (need for a better dosing regimen). JAMA 1991: 265: 361-364.

Automated

Erythrocytapheresis Falciparum

Derek C. Macallan*l, Michaela John Parker-WiIliams2,

6 Steffen R. Fuchs E. Schildknccht J. Mcfloquine compared with other malaria chemoprophylactic regimens in tourists visiting East Africa. J.aizcet 1993: 341: 1299~1303. 7 Kyriacou DN. Spira AM, Talan IX. Mabey DC. Emergency department presentation and misdiagnosis of imported malaria. Am Emerg
in the Treatment Malaria

of Severe

Pocock2, Elizabeth Bishop2, David IH.Bevan2, Tom Harrison’ and Grant T. Robinson2

‘Departnxxt oj- InJxtious Diseasesand ‘Departmerzt

ofHaemntology,

St George’sHospital Medical Scl~ool,Londo~~,U.K.

Removal of parasitized erythrocytes is generally considered to be of value as adjunctive therapy in severe falciparum malaria with high parasitaemia. This is commonly achieved by exchange transfusion. We describe three cases of severe falciparum malaria treated by automated erythrocytapheresis (red cell exchange) in addition to quinine and conventional supportive therapy. Erythrocytapheresis consists of removal of the red-cell fraction by apheresis. Plasma, leukocyte and platelet fractions are returned to the patient. In all cases, dramatic reduction in parasitaemia was achieved within 2 h with subsequent complete clinical recovery. Erythrocytapheresis has significant advantages over exchange ’ transfusion in terms of speed, efficiency, haemodynamic stability and retention of plasma components such as clotting factors and may thus represent an improvement in adjunctive therapy for severe malaria. 0 1999 The British Infection Society

Introduction Palciparum significant

malaria mortality,

with parasitaemia particularly

if there

of more than 10% has a is evidence

of cerebral

involvemenl’,‘. The primary pathology is thought lo be related to sequestration of parasitized red cells in the microcirculation, particularly the cerebral vasculature. Although the mechanism of such oathogenesis is still unclear. manv authorities agree that. at high parasitaemia, removal of parasitized red-cells is an important adjunct to antimalarial chemlotherapy The use of exchange transfusion to achieve this was first described by Gyr et al. in 19743. The rationale behind such an approach is the rapid removal of parasitized cells preventing L

-Address Diseases. Accepted

correspondence to: Dr D. C. Macallan. Division of Infectious St George’s Hospital Medical School. London. SW1 7 ORE, U.K. for publication 10 August 1999.

-

234

ase Reports

further sequestration and improving the rheological properties of blood. Many case reports have now been published describing rapid responses in both clinical and parasitological terms. for example’l. However. i-he place of such therapy and optimum treatment protocols for the management of severe malaria remain contentious’ largely because large controlled studies are lacking. Small studies have shown benefit; for example. a study in Thailand demonstrated a better outcome in four patients treated with exchange transfusion than in 12 who were not”. However. a larger retrospective review failed to show a difference in outcome between units practising exchange and those that did not’. The authors of a recent study of malaria in children in Gabon argued that there was 110 mandate for exchange trausfusion when survival of children with parasitaemias of greater than 10% without such treatment was 98%“. However. children in holo-endemic areas often tolerate very high parasitaemias and it is important not to generalize from such data. Certainly, the associal-ion between high parasitaemia and poor outcome has been recognized for many years] and most studies report significant mortality at levels of parasitaemia above 5% (hyperparasitaemia), particularly in adults returning from visits to an endemic area. Due to the association between parasitaemia and clinical outcome. exchange transfusion has been advocated, where adequate facilities exist. for patients with parasitaemia of more than 15% and clinical complications, or in the 5-15X range if there are other poor prognostic signs.’ Many centres follow this advice. or similar protocols’~“’ even though formal vindication of such an approach in large randomized clinical trials is lacking.’ 1 When a decision is made to treat by exchange transfusion, the procedure is usually performed manually: some authors have recommended continuous exchange. lo whilst others have described either intermittent exchange.” or partial (low volume) exchange’3 I+ to achieve a reduction in the number of parasitized cells. All such approaches are, however. time-consuming. relatively inefficient, and associated with unavoidable haemodynamic disturbance. An alternative approach is described in this paper employing current generation cell separator hardware and software which allows prompt red-cell exchange in a single continuous-flow isovolaemic procedure. precisely tailored to patient size and haematocrit. We describe here the use of automated erythrocytapheresis (red-cell exchange) as adjunctive therapy in three cases of severe falciparum malaria treated in the V.K.

ase

or%!3 Cme 1

A 3%year-old Caucasian woman, normally resident in the U.K.. presented to her local district hospital with a 6-day history of fever and deteriorating mental state. Seven weeks prior to her presentation she had returned from a 12-day holiday in Kenya. She had taken no malaria prophylaxis because of concerns about drug side effects. On admission she was febrile (40. 13C) and confused but was able to obey commands and had a Glasgow coma scale score of 15/I 5. A full blood count showed a haemoglobin concentration of 10.7 g/d], white count of 5.5 x lO’/l and platelets of 14 x 1 P/l. Examination of the blood film showed a 55% parasitaemia with Plasmodium falciparum in which Jnally ring forms but no schizonts were seen. She was hypoglycaemic (glucose 1.6 umol/l) with evidence of renal impairment (Urea 34.8 mmolil. creatinine 326 umol/l, potassium 6.5 mmol/i).

lactic acidosis (lactate 7.X mmol/l). abnormal liver function tests (bilirubin 269 umol/l [normal: O&17]. albumin 28 g/l [38-481) and a C-reactive protein of 199 mg/l[<8]. She commenced treatment with intravenous quinine. receiring a loading dose of 20 mgkg illitiaily followed by 10 mgkg Shourly. She was also given fresh frozen plasma, calcium gluconate and glucose followed by a glucose-insulin infusion to treat hyperltalaemia. She was intubated and ventilated and transferred to the Intensive Care Unit at St. George’s Hospital. Manual exchange transfusion was commenced during transfer. On admission. parasitaemia was found to be 28% and haemoglobin concentration had fallen to 7.1 g/d1 with a platelet count of 10 x 109/1. Two units of plateiets were given to treat thrombocytopenia and cover intravascular canula placement and automated erythrocytapheresis was commenced. A total of 11 units of blood were exchanged. replacing 3 1 of leucodepleted packed cells, over the course of 120 ruin using a Baxter Fenwal CS3000 Blood Cell Separator (Baxter Healthcare Corporation, Round Lake, IL, U.S.A.). The procedure was designed to exchange approximately twice the patient’s red cell volume. As the Cell Separator removes the red cell fraction with a haematocrit of 62% and packed cells have a haematocrit of 63%. a simple volume for volume exchange was performed. No plasma was exchanged. The patient remained %aemodynamically stable throughout the procedure. A net positive balance of 2 units of packed cells was administered in order to maintain haemoglobin concentration (Table I). At the end of apheresis, the parasitaemia was 2.4%. with a haemoglobin concentration of 9.0 g/d1 (Table !). Quinine was continued and parasitaemia continued to fall without additional measures, dropping to ~1% by the second day. Renal support was required for 32 days before recovery of adequate renal funchion. Progress was complicated by hepatic impairment. pneumonia and gastrointestinal haemorrhage secondary to gastric erosions. She was extubated on day 19 and recovered normal mental function. She was discharged ho& well on day 37. Although there was some residual renal impairment at the time of discharge, renal function subsequently returned lo normal.

Case 2 A 26.year-old female LT.:<.-resident presented 20 days following her return from a l-week holiday on the coast of Kenya. She had not taken any anti-malarial prophylaxis. She gave a S-day history of fevers, fatigue and headaches. 3n admission she had a temperature of 39°C but no signs of cerebral involvement Haemoglobin concentration was 12.1 g/d]. white count 6.0 x 109/1 and platelets 23 x 1 O’/l. The blood film showed a 21.5% parasitaemia with Plasmodium falciparum. Rena! function and glucose were normal. She was considered to have ‘severe malaria’ on the basis of her level of parasitaemia despite the absence of cerebral or other organ involvement. She received a loading dose of quinine (20 mgkg) followed by 10 mgkg 8-hourly. She was given 1 unit of platelets to cover line placement and manual exchange transfusion was commenced, removing 3 units of blood and replacing with 4 units of pa&d cells plus fluid replacement. Repeat blood film following this revealed a parasitaemia of 23%. Erythrocytapheresis was therefore commenced, exchanging 11 units of packed cells (2.6 1) over 2 h, as described in Case 1. except that in this case the starting haematocrit was higher, so volume for volume exchange was preceded by the removal of 500 ml of packed cells

235

Case Reports Table I. Effect 01 Erythrocytapheresis on haematological indices and parasilaemia compared to predicted value. Case 1 Haematological indices Haemoglobin concentration (g/dl) Haematocrit Platelets (Xl O”/I) Observed parasilaemia (‘S)” Predicted parasitaemia (‘K)T Exchange parameters Estimated blood volume (1) Estimated erythrocyle volume (1)t Exchange volume (1) Erythrocyte volume exchanged (I)$ Fractional exchange Platelets given (pooled units)6

Pre 7.1 0.198 10 28.0 3.6 0.71 3.0 1.95 2.74 2

Case 2 Post 9.1 0.260 26 2.4 1.4

Pre 12.5 0.362 35 23.0

4.4 1.59 2.8 1.69 1.06

5.4 1.94

0s

Case 3 Post 12.5 0.363 53

2.3 6.4

Pre 12.4 0.339 14 35.0 -

Post 10.8 0.306 42 9.0 9.3

3.5

2.28 1.17 2

Estimated by a single experienced observer, based on reticulocyte counting technique of at leas1 10 fields of 200 cells 7 Calculated as in. I” but modil’ied for erythrocyte exchange only t Eased on estimated blood volume (from standard tables) and measured haematocrit 4:Mean OCremoved and replaced red cell volumes v\Therethese differed: haematocrit of packed cells assumed to be 0.65 4 Number of units of platelets given between pre- and post-procedure results: in Case 2, platelets were given prior to the pre-procedure results shown. 1 unit of platelets represents the pooled platelets from 4 units of blood and contains >240 x 10” platelets in 0.20-0.35 1: one pool of platelets would normally be expected to increase the platelet count in peripheral blood bv about 30-50 x 10”/l

to optimize removal of parasitized red cells. The procedure was uncomplicated except for a short period of hypotension (minimum arterial pressure 70/40). Apheresis was paused for 10 minutes. arterial pressure returned to normal. and apheresis was recommenced without further hypotension. Post-apheresis parasitaemia was 2.3% (Table I). Quinine was continued and she went on to make an uncomplicated recovery.

Case 3

A 3X-year-old man. also UK-resident. presented with a febrile illness 2 weeks after returning from a 4-day visit to Ghana. Prior to this, he had made frequent trips to Ghana on business and did not take any prophylaxis. On the day of admission he had become confused. He was found to have a temperature of 39.6”C and a blood film revealed Plasmodium falciparum with a parasitaemia of 3 5%. Haemoglobin concentration was 13.1 g/dl, white count 3.5 x IO’/1 and platelets 14 x lO’/l. He was treated with intravenous quinine and given 2 units of platelets. Erythrocytapheresis was performed, as described above, without further complications. After removal of 400 ml packed cells, 10 units of packed cells (3.5 1) were exchanged. In this case, leucodepleted cells were not available and hydrocortisone and chlorpheniramine were therefore also given. Immediately postexchange. parasitaemia was 9% and fell thereafter, becoming undetectable by day 4. He developed oliguric renal failure, requiring dialysis from days 3-10. but subsequently made a full recovery.

Discussion These cases demonstrate the use of automated erythrocytapheresis to rapidly and effectively remove parasitized cells in severe falciparum malaria. We have noted an increasing inci-

dence of high parasitaemia falciparum malaria. particularly in travellers taking no prophylaxis, at our reference centre, which suggests that such management may become increasingly important. The use of erythrocytapheresis to remove pathological redcells has been well-described in the analogous haemorheological emergency of life-threatening sickle cell crisis.” “.lT Its use in malaria has only once before been described to our l 12.0 gidl).” No other major complications were observed. The advantages of automated erythrocytapheresis over manual exchange are efficacy, speed, haemodynamic control and retention of plasma and cellular components. In terms of efficiency. post-procedure parasite counts were ~3% in two cases and close to predicted values, assuming a single compartment exchange model modified from that described by Wilkinson et al.‘O (Table I). Automated exchanges were completed within 2 h, whereas standard exchange transfusion of 11 units would tale over 5 h at the recommended rate of 500 ml exchanges over 20-30 minutes.‘O Haemodynamic disturbance was minimized by the removal and replacement of red cells at the same haematocrit and rate and the potentially deleterious effects of fluctuations in volume12 were thus avoided. Hypotension did occur briefly in one case (2) but it was not clear whether this was related to the procedure, particularly since isovolaemia had been maintained throughout. The primary disadvantage of the procedure is the requirement for appropriate equipment and a trained operator.

In these cases of severe falciparum malaria. the use 0r automated erythrocytapheresis was associated with a prompt reduction in parasitaemia (Fig. 1) and subsequent MI recovery The procedure was well tolerated. Where a decision has been made bo institute exchange transfusion in severe malaria. we believe that erythrocytapheresis should be considered in preference to conventional manual exchange transfusion because of the potential advantages in terms of efficacy speed, haemodynamic control and retention of plasma and cellular components. 20

1 4 0

12

Time(h)

after

1 Parasitaemia following the three cases described.

Figure

24

36

48

60

erythrocytapheresis ergthrocylapheresis

(shaded

area) for

Although unfamiliar to many Infectious Disease units. cell separator hardware is likely lo be available in many Haematology l~epartments

in

industrialized

countries.

However.

this

does

mean thai the use of erythrocytapheresis will be restricted to those centres where such facilities are available and it is unlikely to be widely

applicable

to deveieloping

world

settings.

Assuming

hardware purchase costs have been covered. the cost per procedure is about f220. not including the cost of blood products. As with manual exchange transfusion. liypocalcaeiuia is a potential problem due to citrate in transfused blood and calcium infusions are usually given routinely to prevent this. In erythrocytapheresis, plasma is returned to the palient and erythrocytapheresis is therefore kly to cause less metabolic, immune.

coagulation

and

pharnlacokinetic

disturbance

than

exchange transfusion. By con?rast, some authorities have advocated plasmapheresis as treatment for severe malaria.‘” This, ii. is argued. removes ‘toxins’, such as parasite products. immune complexes and pro-inflammatory mediators. We feel that with erythrocytapheresis, the advantages of retaining beneficial plasma components such as clotting factors far outweigh the theoretical disadvantages of retaining putative toxic factors. Indeed some pro-inflammatory mediatcrs may have a beneficial protective effect themselves. In addition, the leultoc:7te aild platelet fractions are largely retained by the procedure with obvious theoretical benefit. Platelet loss during erythrocytapheresis is minimal as platelets are hypodense relative to huffy coat and are returned to the patient with the plasma. In ali three cases described above. platelets were given prior to the procedure, primarily to prevent bleeding related to intravascular canula placement. In Case 2. no platelets were given between the pre- and post- procedure blood counts and ii. can be seen from Table I that the platelet count was well maintained and no1 depleted by apheresis. Neither erythrocytapheresis nor conventional exchange transfusion are thought to remove parasitised erythrocytes which have already sequestered. However. both treatments may prevent further sequestration and improve flow in partially obstructed vessels by improving the rheological characteristics of circulating blood.

10

11 12

13

14

15

16 17 18

!9

Field JW Blood examination and prognosis in accle Ea!ciparum malaria. Tr~ns lzoy Sot Z?ophlrdHyg 194Y; 43: 33-48. Miller KD, Greenberg AE. Campbell CC. Trcalment of severe malaria in the IJniled Slates with a continuous infusion of quinidine gluconale and exchange lransfusion. NE~7~~~JUeii 7 989: 321: 65-70. Gyr I<. Speck E, Rilz R. Cornu II, Buckner CD Zerebralle malaria trap ica mit Scilr~arzwasserfieber. Schuviz ,‘vl~d ilkthenschr J 9 74: 104: 1628-1630. Looareesuwan S, Phillips RE. I
0r