Retrospective Immunodiagnosis of Malaria in Nonimmune Travelers Returning From the Tropics TomasJelinek, Frank von Sonnenbuty, Susanna Kumlien, Thomas Loschev, and Hans D. Nothduft

Background: Since travelers are now frequently advised to carry emergency self medication for the treatment of sus- pected malaria, the issue of retrospective diagnosis of malaria in nonimmune patients has recently gained more importance. Reliable methods for evaluating the frequency and justifying the use of self medication in clinically suspected malaria are warranted.

Methods: One hundred and eighty-five sera from 132 nonimmune travelers returning to a nonendemic area, and with a microscopically confirmed diagnosis of malaria, were investigated for the presence of antibodies against blood stages of Plasmodium fakiparum or Plasmodium vivax, by the indirect fluorescence antibody test (IFAT). Eighty-eight patients suffered from infection with /? falciparum and 44 from the infection with I? vivax. In fakiparum malaria, 97.1% of patients had positive reactions during a period of 15-60 days after onset of symptoms. During the same period after onset of symptoms, significant titers were demonstrated in 88.2% of patients with vivaxmalaria. Extended cross-reactions between the antigens used and a wide range of interindividual differences in antibody titers were observed. One hundred sera from Germans recovering from nonmalarial febrile illnesses were used as negative control group in the investigation, of which none resulted in a positive IFAT.

Conclusions: We conclude from these results that the IFAT is a specific and sensitive tool for the retrospective confirma- tion of malaria in the differential diagnosis of fever imported from endemic areas by nonimmune travelers. Nevertheless, when dealing with the individual patient, a careful interpretation, with inclusion of al l available clinical data, is manda- tory. However, by using blood stage antigens, the IFAT can be considered a sensitive tool in epidemiologic surveys. The tool can be used with a high degree of reliability, even without access to additional clinical data.

Following the increase of international air travel in recent years, malaria has become a diagnostic and thera- peutic problem in travelers returning from tropical coun- tries.The disease has to be considered in all febrile patients with a recent history of travel to endemic areas. The serologic diagnosis is of no particular value in acute malaria since the fast and simple tools ofthe thin and thick blood film are available. However, the detection of malaria antibodies can be useful for the retrospective diagnosis of malaria when no parasites are detectable in the blood.This might be especially important in patients presenting after a journey to endemic areas who report an episode of fever that arouses a suspicion of malaria that was never confirmed by microscopy. Since travelers are now frequently advised to carry emergency self medication for the treatment of suspected malaria, this issue has gained more importance. Also, an infection with l? vivax or Plasmodium ovale has to

T. Jelinek, MD, Ev. Sonnenburg, MD, S. Kumlien, MSc, T. Loscher, MD, and H.D. Nothdun?, MD: Department of Infectious Diseases and Tropical Medicine, University of Munich, Munich, Germany

Reprint requests: Tomas Jelinek, MD, Department for Infectious Diseases and Tropical Medicine, Leopoldstrasse 5, 80802 Munich, Germany

J Travel Med 1995; 23225228.

be verified since, in order to eradicate the liver stages of the parasite, treatment has to be completed by a course of primaquine. In this study, we attempted to demonstrate the reliability of the most commonly used serologic method, the indirect immunofluorescence antibody test (IFAT), and its value for the retrospective diagnosis of malaria. We focused our attention on nonimmune trav- elers since this group has been underrepresented in pre- vious investigations.'-'

Methods

One hundred and eighty-five specimens from 132 patients with malaria, previously diagnosed in our out- patient clinic over a period of 5 years, were investigated in this retrospective study. All patients were German nationals or residents for more than 10 years. All speci- mens derived after a reexposure to endemic areas (and therefore possible reinfection with malaria) were excluded from the investigation.The time of exposure to malaria in endemic areas varied from 6 days to 2.5 years, with a median of 28 days. In all cases, the diagnosis was con- firmed microscopically by thin and thick blood films. In all patients, treatment was started following microscopic confirmation of the diagnosis. None of the patients had been treated prior to presentation at the clinic.

Specimens were derived during the first presenta- tion and various follow ups.The longest period between

225

2 2 6 Jou rna l o f T rave l Medicine, Vo lume 2, Number 4

Table 1 Results of Indirect Fluorescence Antibody Test (IFAT) in Confirmed Cases of Malaria in Nonirnrnune Travelers

Antigen Infecting Parasite and Time Since Onset of Symptoms

Pfakiparum (n = 118 Specimens Derived From 88 Patients)

(37 Days 8-14 Days 15-60 Days 61-180 Days >I81 Days (n=47) (n=16) (n =3 4) (n=15) (n=6)

I? falciparum Positive ('%I)* 13(27.7) 14(87.5) 33(97.1) 11 (73.3) l(16.7)

I? vivax Positive ('h) h(12.7) 6(37.5) 2 1 (6 1.8) 8 (53.3) l(l6.7) GMT** 10 1 ( 2 5 1.8) 237( ? 86.5) 347.3(?52.1) 228( 586.1) 26.7(? 15.9)

GMT 36.4(?31.4) 78(?63.1) 73.9(523.4) 90.7(?61.4) 10.7(119.1)

I? vivax (n = 67 Specimens Derived From 44 Patients)

61-180 Days >I81 Days (n = 16) (n = 7) (n = 17) (n = 10) (n = 17)

0-7 Days 8- 14 Days 15-60 Days

I?falciparum Positive ("h) 8(50) 4(57.1) 1 O(58.8) 7(70) 8(47.1)

L? vivax Positive ("h) 8(50) 7(100) 1 5 (88.2) lO(100) 14(82.4) GMT 1 20.5 ( ? 80.5)

GMT 212(?103.7) 365.7(?128.5) 320.9(5103.7) 396.8(?90.1) 232.5(594.8)

140(?93.4) 1 7 6 ( 1 1 5 7.9) 1 37.4 ( ? 77.1 ) 1 32.8 ( ? 89.8)

*Positive titer 21 54; **GMT, geometric mean titer (95% confidence interval)

onset of symptoms and last follow up was 812 days in a patient withfakiparum malaria and 1825 days in a patient with vivax malaria, respectively.The sera were examined in the dilutions 1:16,1:32,1:64,1:128,1:256 and 1:512 by the indirect immunofluorescence antibody test (IFAT). Titers from 1 :64, and higher, were considered positive as suggested previously in the literature.s" Mature tropho- zoites and schizonts of Pfalciparnm derived from con- tinuous culture5." and of P vivax obtained from infected individuals were used as antigens.'","

One hundred sera from 100 Germans recovering from nonmalarial febrile illnesses were used as a nega- tive control group in the investigation.These patients were matched according to travel destination and age.The sera were derived during routine examinations from the out- patient clinic. Seventy-eight percent of these patients suf- fered from flu-like symptoms, including fever, without detection of any pathogen; 13% from other unspecified viral infections; 5% from bacterial dysentery; and 4% from amebic dysentery.

Data were entered into a computerized data base and analyzed using chi-square tests to determine statistical dif- ferences. Statistical significance between the median titer steps (GMT) was calculated with a paired t test.

Results

Altogether we analyzed 185 single sera from the 132 patients under investigation (Table 1). Eighty-eight patients suffered fromfalciparurn malaria and 44 patients from vivax malaria.AU patients exhibited clinical malaria

at the first presentation. In most patients, it was only pos- sible to obtain one specimen. Serial observations with investigation of several specimens could be made in 27 cases (18 with P falciparum and 9 with P vivax). Signifi- cant antibody titers were detectable in 97.1% of the patients withfaleiparurn malaria during a period of 15-60 days after onset of symptoms (see Table I). A steady decrease of the antibody titers could be observed in sera derived later. In vivax malaria, all patients had significant antibody levels from the eighth day after onset of symp- toms. In this form of malaria, prolonged antibody response was notable. In both types of infection, almost all patients developed significant antibody titers during days 8-1 4 after onset. Large interindividual differences in titer lev- els after infection were apparent.These varied from 1:0 to 1 :512 or higher. Considerable cross-reactivity between P-fakiparum and I? vivax was frequent.

Serial observations were done in 18 patients with faleiparurn malaria and 9 patients with vivax malaria.The results of the evaluation of the serologic data alone were reflected in both groups. In thefaleiparurn group, 17 out of 18 patients showed positive antibody titers within 14 days after onset of symptoms.The geometric mean titer (GMT) was 244 +- 114.6 (95% confidence interval) for the period of 0-14 days.The patients in the vivax group showed a somewhat delayed reaction (six out of nine positive within the first 14 days, GMT 256 2 179 [95% confidence level]), but within 15-60 days, after onset of symptoms, significant titers could be demonstrated in all cases with a GMT 387 % 153 (95% confidence interval).

J e l i n e k , S o n n e n b u r g , e t a l , R e t r o s p e c t i v e l r n r n u n o d i a g n o s i s o f M a l a r i a 2 2 7

All 100 sera from patients with illnesses other than malaria were negative in the IFAT. The highest titer obtained was 1:32 in three cases; the GMT was located at 4.5 -+ 13.8 (95% confidence interval).

Discussion

When analyzing the results, although hampered by a possible bias attributable to interindividual differ- ences, the IFAT has been described as the most specific and sensitive method for the detection of antibodies against malaria parasite^.^^'^^^'^'^^^' O ur results indicate that significant antibodies against malaria parasites can be detected in a majority of patients from the eighth day after onset of symptoms by IFAT. The titer seems to depend on the amount of antigenic stimulus, which is related to the species of parasite, the occurrence of relapses, and the immunologic capability and memory of the host.This assumption could easily explain the high differences in titer levels between individuals during the same period after onset of symptoms. Similar results were reported previously.'

From the 60th day onwards after onset of symptoms, antibody levels showed a steady decrease in those patients with falciparum malaria. Patients infected with I? vivax showed a delayed reaction with regard to the decrease of antibody response. Retrospective diagnosis should be possible in a high percentage of cases with falciparum malaria, even after more than 1 year following the onset of symptoms.The reasons for the markedly slower decline of antibody titers in vivax malaria are not fully under- stood. One reason might be that vivax malaria, unlikefal- ciparum malaria, frequently leads to recurring attacks and therefore to increased antigen presentation in nonim- munes.The antigen preparation for the IFAT might also play some role (cultured I? fakiparum parasites and I? vivax derived from infected patients).This might well be a rea- son for the faster decline of antibody response in patients with falciparum malaria in the investigated population.

Using both Pfakiparum and I? vivax antigens can help to differentiate between malarial species by determina- tion of the highest titer. This feature can be useful in the identification of patients with vivax malaria where there is a need for a therapeutic course of primaquine, even for a considerable time after infection. However, the high amount of cross reactions, and the large inter- individual titer variations, as observed in our investiga- tion, has to be kept in mind (seeTable 1).

The investigation of the negative control group demonstrated a specificity of 100% for the test method. Therefore, the retrospective diagnosis of a malaria infec- tion in nonimmune travelers, without previous malarial infection, can be obtained with some certainty in the case of a positive test result.

Without clinical data, the high degree of inter- individual variability of antibody titers makes it dificult to determine the specific time of malaria infection. Usu- ally it can only be stated that a clinical attack has occurred. This is because it can take some time to develop high titers, and these usually persist for many months.This is especially true for I? vivax infection^.^.^.^

Previously a connection between a weak antibody response and early treatment of malaria has been postu- 1ated.This connection could neither be confirmed nor rejected with our data because the sample size in the group, and the serial observations, was too small to cal- culate significant statistical differences. However, the courses of antibody response in patients treated within 2 days after onset of symptoms, and treated up to 12 days later, showed no marked differences.

Previous investigations showed that the interpreta- tion of antibody titers against malaria parasites in per- sons living in endemic areas is dificult.'J2J3 Positive titers may indicate a recent acute'attack, a low grade asymptomatic infection, or a persisting titer after a pre- vious infection. In nonimmune subjects, however, the serologic diagnosis of malaria has an important value for the retrospective differential diagnosis of fever acquired in tropical areas. A careful interpretation with inclusion of all available clinical data is necessary.Therefore, when dealing with the individual malaria patient, the inter- pretation of serologic data can be difficult.This is espe- cially true if clinical decisions are purely based on serologic results. Nevertheless, the IFAT is a valuable tool for epi- demiologic surveys in malaria. Its high sensitivity in the period of 15 to 60 days after onset of symptoms should lead to high reliability of data obtained during this time.

References

1.

2.

3.

4.

5.

6.

Draper CC, Sirr SS. Serological investigations in retrospective diagnosis of malaria. BMJ 1980; 1575-1576. Kanol-Ratanahul P, Chirahalwasaru N, Lertmaharit S, Dhanamun 13, et al. Seroepidenmiologic studies of humoral immune response to the r? falripamm antigens inThailand. Am JTrop Med Hyg 1992; 47:554-561. Mvondo JL, James MA, Sulzer AJ, Capbell CC. Malaria and pregnancy in Cameroonian woman.Trans R SocTrop Med Hyg 1992; 86:486-490. Sutanto I , Pribadi W, Muljono R, Susanto L, et al. Sero- epidemiological studies of malaria in different endemic areas of Indonesia. Southeast Asian J Trop Med Public Health

Hall CL, Haynes JD, Chulay JD, Diggs CL. Cultured P fakiparum used as antigen in a malaria indirect fluorescent antibody test. Am JTrop Med Hyg 1978; 27:849-855. Manawadu BR,Voller A. Detection and measurement of species-specific malarial antibodies by immunofluorescence tests.Trans Roy SocTrop Med Hyg 1978; 72:463-466.

1992; 23:563-569.

2 2 8 Journa l of Travel Medicine, Volume 2, Number 4

7.

8.

9.

10.

Manawadu BR, Voller A. Standardization of the indirect fluorescent antibody test for malaria.Trans Roy Soc Med Hyg

Wilson M, Sulzer AJ, Runcik K. Malaria antibody patterns as determined by the IFA test in U.S. serviceman after chemotherapy. Am J Trop Med Hyg 1970; 19:401-404. Trager W, Jensen JB. Human malaria parasites in continuous culture. Science 1976; 193:673-674. Sulzer A,Wilson M,TurnerA, Kagan 1.A multispecies malaria antigen for use in the indirect fluorescent antibody tect.Trans Roy SocTrop Med Hyg 1973; 67:55-58.

1978; 72~456-462.

11. Janse CJ, Carmago A, Del Portillo HA, Herrera S, et al. Removal of leucocytes from Plasmodium vivax-infected blood. AnnTrop Med Parasitol 1994; 88:213-216.

12. Bruce-Chwatt LJ, Dodge JS, Draper CC,Topley E, et al. Sero- epidemiological studies on population groups previously exposed to malaria. Lancet 1972; 1:512-515.

13. Collins WE, SkinnerJC,Jeffet-y GM. Studies on the persistence of malaria antibody response. A m J Epidemiol 1968; 87:592-598.

Wat Chai Mongkul, Ayutthaya,Thailand. Submitted by Christoph Hatz, MD.