Patient‐to‐Patient Transmission of Nosocomial Malaria in Italy • Author(s): Maria Luisa Moro , MD; Roberto Romi , PhD; Carlo Severini , PhD; Gian PaoloCasadio , MD; Giovanni Sarta , MD; Guido Tampieri , MD; Antonio Scardovi , MD; CinziaPozzetti , RN, CIC; the Malaria Outbreak GroupSource: Infection Control and Hospital Epidemiology, Vol. 23, No. 6 (June 2002), pp. 338-341Published by: The University of Chicago Press on behalf of The Society for Healthcare Epidemiologyof AmericaStable URL: http://www.jstor.org/stable/10.1086/502062 .

Accessed: 14/05/2014 12:42

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact support@jstor.org.

.

The University of Chicago Press and The Society for Healthcare Epidemiology of America are collaboratingwith JSTOR to digitize, preserve and extend access to Infection Control and Hospital Epidemiology.

http://www.jstor.org

This content downloaded from 195.78.109.14 on Wed, 14 May 2014 12:42:47 PMAll use subject to JSTOR Terms and Conditions

338 INFECTION CONTROL AND HOSPITAL EPIDEMIOLOGY June 2002

PATIENT-TO-PATIENT TRANSMISSION OF NOSOCOMIAL

MALARIA IN ITALY

Maria Luisa Moro, MD; Roberto Romi, PhD; Carlo Severini, PhD; Gian Paolo Casadio, MD; Giovanni Sarta, MD; Guido Tampieri, MD; Antonio Scardovi, MD; Cinzia Pozzetti, RN, CIC; the Malaria Outbreak Group

OBJECTIVE: To describe nosocomial transmission ofmalaria from patient to patient via blood exposure.

PATIENTS: A 56-year-old man was admitted to an Italianhospital with fever and Plasmodium falciparum parasitemia, butwith no risk factors for malaria. Twenty days earlier, he had beenadmitted for bronchopulmonary disease to the hospital’s intensivecare unit, where a woman with P. falciparum malaria acquiredabroad was present.

METHODS: We reviewed both patients’ medical recordsand searched for mosquitoes in the hospital and on the grounds.We interviewed the staff about patient care practices potentiallyinvolving contact with blood. The genetic identities of strains weredetermined by genotyping of the DNA extracted from blood.

RESULTS: Molecular genotyping showed that the twostrains were identical. The only invasive procedures performed onboth patients by the same staff on the same shift were capillary

blood sampling by finger stick, intravenous drug administration,and substitution of total parenteral nutrition bags and intravenoussets. The fingerstick device used was designed to prevent person-to-person transmission of blood-borne infections, and the staffinterviews did not reveal any incorrect use of aseptic techniques.The likely source of infection was identified during a trainingcourse 6 months later: a nurse reported that, when collectingblood, she placed patients’ fingers directly on the blood glucosemeter, a practice she had learned from a poster advertising thedevice.

CONCLUSIONS: A nosocomial case of malaria was ascer-tained, which was likely due to patient-to-patient transmission via acontaminated blood glucose meter. Incomplete instructions for themeter seem to have played a role in this case (Infect Control HospEpidemiol 2002;23:338-341).

Although nosocomial transmission of malaria is rarein nonendemic areas, patient-to-staff transmission byneedlestick injury or by contact with the blood of openwounds has been documented, as has patient-to-patienttransmission through organ transplants.1 There have alsobeen reports of four outbreaks2-6 and a single case7 of trans-mission of malaria associated with contaminated multidosevials or intravenous solutions.

In Italy, malaria was eradicated nearly 50 years ago.Since then, a national malaria surveillance system has beenresponsible for detecting locally acquired cases. Malaria iscurrently the most common imported infectious disease inItaly. From 1990 through 2000, only 21 (0.2%) of the 8,153confirmed cases, including the 1 described in this study,were classified as autochthonous8,9: 8 were caused by bloodtransfusion; 1 occurred in a drug addict; 1 occurred in aphysician through needlestick injury8; 2 were airportmalaria; 7 were baggage malaria; and 1 was transmitted by

an indigenous mosquito, the first after eradication of malar-ia.10

In this report, we describe a case of nosocomialmalaria acquired in an intensive care unit for which person-to-person transmission was supported by molecular typing.

METHODS

Case ReportOn April 1, 2000, a 56-year-old man was admitted

with fever and Plasmodium falciparum parasitemia to apublic general hospital in northern Italy. He was trans-ferred to the intensive care unit because of acute renal fail-ure and disseminated intravascular coagulation. Neitherthe man nor his family had traveled to areas endemic formalaria, nor had he been living near a harbor or an airport.The man had never undergone transfusion of blood orblood products, and he reported that he had never usedintravenous drugs. Transmission of malaria had never

Dr. Moro is from the Laboratorio di Epidemiologia e Biostatistica, Istituto Superiore di Sanità, Rome, Italy. Drs. Romi and Severini are from theLaboratorio di Parassitologia, Istituto Superiore di Sanità, Rome, Italy. Dr. Casadio is from the Dipartimento di Prevenzione, Lugo AUSL Ravenna,Italy. Dr. Sarta is from the Laboratorio di Microbiologia, Ospedale di Lugo, Ravenna, Italy. Dr. Tampieri is from the Servizio Emodialisi, Ospedale diLugo, Ravenna, Italy. Dr. Scardovi is from the Centro di Rianimazione, Ospedale di Lugo, Ravenna, Italy. Mrs. Pozzetti is from the Direzione Sanitaria,Ospedale di Lugo, Ravenna, Italy. The Malaria Outbreak Group consists of Michela Menegon, Marco Di Luca, Valeria Contarini, Claudio Assirelli,Andrea Sangiorgi, Giovanni Sarta, Guido Tampieri, Antonio Scardovi, Antonio Zappi, Gabriella Dovadola, Mauro Zaffagnini, Augusto Mignani,Davide Tellarini, and Lorella Ricci.

Address reprint requests to Dr. Maria Luisa Moro, Agenzia Sanitaria Regionale, Area di Programma Rischio Infettivo, Via Gramsci 12, 40121Bologna, Italy.

ABSTRACT

This content downloaded from 195.78.109.14 on Wed, 14 May 2014 12:42:47 PMAll use subject to JSTOR Terms and Conditions

Vol. 23 No. 6 NOSOCOMIAL TRANSMISSION OF MALARIA 339

occurred in the area where he had been living, and nocases of malaria had been documented in this area in theprevious 6 months.

The man had previously been admitted for bron-chopulmonary disease to the same hospital, specifically tothe intensive care unit from March 6 to March 10 (until 9am), 2000, and to the respiratory unit from March 10 toMarch 23, 2000. From March 7 to 16, 2000, a 51-year-oldwoman who had been admitted for P. falciparum malariaacquired in Africa (22% of infected erythrocytes on March7) was in the same intensive care unit.

Procedure ReviewWe reviewed the medical records of the man and the

woman. Data were collected regarding malaria species andparasitemia, location within the intensive care unit, specifictype of blood work and invasive procedures, intravenousdevices and medications (ie, the specific device used, thefluids and medications received, and the time of manipula-tion), other types of percutaneous exposure (eg, fingerstickdevices), and administration of blood products.

We questioned the hospital staff about the presenceof mosquitoes in the hospital. We thoroughly inspected theintensive care unit for the presence of mosquitoes and setlight traps on the grounds just outside the intensive careunit. We also interviewed the staff regarding the methodsused to prepare and administer intravenous medication andthe manipulation of intravenous devices. The fingerstickdevice used in the intensive care unit was inspected andstaff were interviewed about its manipulation.

Molecular GenotypingWe ascertained the genetic identities of the two plas-

modial isolates involved in the nosocomial transmission bygenotyping the parasite DNA extracted from the blood ofthe two patients. Five blood specimens infected with P. fal-ciparum were used as controls: two laboratory strains (3D7and 1776S8, currently maintained in continuous culture inthe Istituto Superiore di Sanità laboratory) and three clini-cal isolates (Italian cases imported from Senegal, Guinea,and Zambia). Genomic DNA was extracted from 1 mL offrozen blood for both the man and the controls and fromblood smears for the woman because she no longer hadparasitemia when the investigation began. All DNA extrac-tions were performed using the Qiagen DNA Easy Kit(Qiagen Inc., Valencia, CA), following the manufacturer’sinstructions. The DNA was purified and concentrated usingthe Microcon-100 system (Millipore Corp., Bedford, MA).The following four molecular markers were selected andamplified: merozoite surface protein 1 (MSP1) and mero-zoite surface protein 2 (MSP2)11; P. falciparum gametocyte377 (Pfg 377)12; and glutamate-rich protein (GLURP).13

Polymerase chain reaction (PCR) products were directlysequenced using the ABI PRISM Big Dye TerminatorCycle Sequencing Ready Reaction Kit (Applied Biosystems,Inc., Foster City, CA), following the manufacturer’s instruc-tions. Nucleotide sequences were analyzed by an automat-ic sequence reader (Applied Biosystems, Inc.) and com-

pared using the PC/GENE program (IntelliGenetics Inc.,Mountain View, CA).

RESULTS

No mosquitoes were detected in the intensive careunit, but a total of 8 mosquitoes were detected on thegrounds outside the intensive care unit and in the man’shome; however, they did not belong to genera capable oftransmitting malaria (7 Culex pipiens and 1 Aedes genicula-tus). When admitted to the hospital, the index case had notbrought any luggage capable of carrying mosquitoes. Thepeak daily temperatures recorded outside the hospital inthe first 10 days of March ranged from 12.3°C to 22.8°C(median, 16.8°C).

During the 3 days that the two patients had bothbeen in the intensive care unit (ie, March 7, 8, and 9), theywere in separate rooms across the hall from each other.They both had a central venous catheter in place andreceived multiple intravenous drug administrations.Neither of them underwent transfusion of blood or bloodproducts. Arterial blood gases and blood sampling wereperformed for both patients on 1 day only, but by differentstaff on separate shifts. Multidose vials were not used foreither of the patients; the heparin used for flushing the cen-tral venous catheters was supplied by the hospital pharma-cy as unit doses. According to the staff, syringes had notbeen used on more than one patient. No radiographicexamination with contrast medium was performed oneither of the patients.

The only invasive procedures performed on bothpatients by the same staff on the same shift on at least 1 ofthe 3 days that they were both in the intensive care unitwere capillary blood sampling by finger stick (five differentoccasions), intravenous administration of drugs (nine occa-sions), substitution of the total parenteral nutrition bag(one occasion), and substitution of the intravenous set (oneoccasion). The fingerstick device used in the intensive careunit (Softclix-Pro, Roche, Basel, Switzerland) is designed toprevent accidental person-to-person transmission of blood-borne infections; specifically, it does not function if both theneedle and the platform are not changed between uses. Allof the nurses were also interviewed about the use of theblood glucose meter (Glucotrend, Roche), and all of themreported that, when testing blood, they first removed thestrip from the meter. None of the five other patients stayingin the intensive care unit during the same 3 days shared bythe source case and the secondary case of malaria hadtheir blood glucose levels measured.

On the basis of these results, it was not possible toidentify the source of infection. We thus assumed that theinfection was due to the use of either the same syringe onthe two patients when flushing the intravenous catheters(despite the provision of heparin in unit doses) or contami-nated gloves when handling tools that came into contactwith blood (eg, the fingerstick device or the central venouscatheter). Six months after the occurrence of the case ofmalaria, a training course was held to reinforce the infec-tion control measures adopted in the intensive care unit,

This content downloaded from 195.78.109.14 on Wed, 14 May 2014 12:42:47 PMAll use subject to JSTOR Terms and Conditions

340 INFECTION CONTROL AND HOSPITAL EPIDEMIOLOGY June 2002

which is especially important given the high turnover ofnursing staff. During the course, one of the nurses report-ed that, when testing blood, she placed patients’ fingersdirectly on the meter and that this practice was recom-mended on a poster about how to use the device that hadprobably been supplied by a pharmaceutical company rep-resentative. We obtained a copy of the poster and foundthat the instructions specified that the blood should beapplied to the test strip by placing the patient’s finger eitherdirectly on the strip, without removing it from the meter, oron the strip after having removed it from the meter. Theinstruction sheet supplied by the manufacturer correctlystated that the first option should be reserved for single-patient use, but the poster did not.

Molecular GenotypingSpecific portions of the genomic DNA of parasites

were amplified with primers relevant to the polymorphicregions of the genes codifying the surface antigens of theP. falciparum. In particular, the following targets wereselected: a portion of 200 to 300 nucleotides of the MSP111;a portion of 400 to 500 nucleotides of the MSP211; region“3” of the Pfg 37712; and a portion of 900 to 1,000nucleotides of the GLURP.13 The amplification products rel-evant to the two clinical isolates were identical in size to thefive reference isolates. The comparison of the nucleotidesequences of the amplicons relevant to the two clinical iso-lates demonstrated a 100% identity for each of the four

polymorphic genes considered. By contrast, none of thethree clinical control isolates showed 100% identity witheither of the two isolates from the two patients in this studyusing any of the four markers. The two laboratory controlstrains each showed 100% identity using one of the fourmarkers with both of the isolates from the two patients inthis study; this 100% identity was demonstrated usingGLURP for laboratory control strain 3D7 and using MSP1for laboratory control strain 1776S8 (Table).

DISCUSSION

All of the reported cases of patient-to-patient trans-mission of malaria implicated solutions, medications, orcontrast medium contaminated with the blood of patientsinfected with malaria and directly infused to secondarycases.2-6 In our case, no exposure to contaminated intra-venous solutions, catheters, or syringes was identified.Nonetheless, several findings strongly support the hypoth-esis of nosocomial transmission, the most important ofwhich was the finding that the two strains were geneticallyidentical. Some relatively recent studies have used PCRamplification for the genotyping of malaria parasite popula-tions.11,13 This technique, which is reproducible and sensi-tive,11,13 has greatly simplified the analysis of allelic diversi-ty in P. falciparum. It is also commonly used in clinical studies aimed at distinguishing new infections from recur-rences.14 The genes coding for highly polymorphic anti-gens are commonly used as genetic markers because, as

TABLERESULTS OF THE GENOTYPING ANALYSIS OF THE TWO ISOLATES INVOLVED IN THE NOSOCOMIAL CASE OF MALARIA AND OF THE FIVE

CONTROL SPECIMENS*

Isolates Involved inthe Nosocomial Case

Source Secondary Control Specimens(Woman) Case (Man) 3D7 1776S8 Senegal Zambia Guinea

Source MSP1, GLURP MSP1(woman) MSP2,

Pfg 377, GLURP

Secondary MSP1, GLURP MSP1case (man) MSP2,

Pfg 377,GLURP

3D7 GLURP GLURP MSP1 MSP1, MSP1Pfg 377

1776S8 MSP1 MSP1 MSP2 Pfg 377 Pfg 377Senegal MSP1 MSP2 MSP1 MSP1,

GLURPZambia MSP1, Pfg 377 MSP1 MSP1,

Pfg 377 Pfg 377Guinea MSP1 Pfg 377 MSP1, MSP1,

GLURP Pfg 377

MSP = merozoite surface protein; Pfg = Plasmodium falciparum gametocyte 377; GLURP = glutamate-rich protein.*The molecular markers showing 100% identity at the pair-wise comparison are listed.

This content downloaded from 195.78.109.14 on Wed, 14 May 2014 12:42:47 PMAll use subject to JSTOR Terms and Conditions

Vol. 23 No. 6 NOSOCOMIAL TRANSMISSION OF MALARIA 341

recently reported in the literature, the identical ampliconsize and nucleotide sequence of at least three of such poly-morphic markers provide 90% identity.13-15 Moreover, byincreasing the number of genes included in the analysis,and if these genes have an increased level of polymorphismin the target regions, the likelihood of accurate identifica-tion increases. In our study, four polymorphic markerswere analyzed, and an identical genotype pattern was foundfor the two isolates, indicating that the same P. falciparumgenotype was probably transmitted from the index case tothe other patient.

Although we cannot completely exclude the alterna-tive hypothesis of hospital-acquired, mosquito-borne malar-ia, this is unlikely because we found no mosquitoes eitherinside or outside the hospital, the environmental conditionswere not favorable for the presence of Anopheles, and theindex case did not bring any luggage to the hospital, apartfrom a small cosmetics case. It is much more likely that theinfection had been transmitted through contact with conta-minated blood, probably by capillary blood sampling forblood glucose testing. In fact, several outbreaks of blood-borne infections (mainly hepatitis B virus infection) havebeen associated with reusable fingerstick blood-samplingdevices.16-18

Recognizing this risk, the infection control team ofthe hospital had selected a capillary blood-sampling devicethat was considered to be safe, yet transmission occurred,probably through a contaminated blood glucose testingdevice. This cannot be definitely proved, but it is highlylikely: blood glucose testing was the only procedure sharedby the two patients, a young nurse (not involved in thecase) admitted that this practice was in use, and this hasbeen confirmed by other nurses as well. No other case ofblood-borne infection was identified among the patientsstaying in the intensive care unit, but it cannot be ruled outbecause no active case finding has been performed.

This case of nosocomial transmission is a goodexample of how several factors, which are sometimes diffi-cult to foresee, can influence the effectiveness of infectioncontrol measures. Although articles have discussed theimpact of inaccurate messages provided by pharmaceuticalcompany representatives on the prescribing habits ofphysicians,19 the impact of nonstandard patient care prac-tices due to inaccurate informational materials provided bythe manufacturers of specific devices has not been ade-quately evaluated. This important issue should be consid-ered by infection control teams, who should work togetherwith representatives from device manufacturers, regulato-ry agencies, and the scientific community to ensure thatthe information provided in promotional messages is basedon sound scientific evidence and that these messages areclear and accurate.

Another factor that must be considered in investigat-ing such outbreaks is the difficulty in identifying incorrectpatient care practices. In our study, despite the fact that weconducted thorough staff interviews, the incorrect use ofthe meter was revealed 6 months later and only by chance.

This case should also serve as a reminder that therisk of malaria exposure in hospitals located in nonendem-ic areas is increasing as a result of increased travel to andfrom endemic areas, and that strict aseptic proceduresshould be rigorously applied to prevent the transmission ofpotentially severe infections.

REFERENCES1. Lettau LA. Nosocomial transmission and infection control aspects of par-

asitic and ectoparasitic diseases: Part II. Blood and tissue parasites.Infect Control Hosp Epidemiol 1991;12:111-121.

2. Jensen JB, Capps TC, Carlin JM. Clinical drug-resistant falciparummalaria acquired from cultured parasites. Am J Trop Med Hyg 1981;30:523-525.

3. Anonymous. Hospital-acquired malaria in Nottingham. Commun Dis RepCDR Wkly 1999;9:123.

4. Chen K-T, Chen C-J, Chang P-Y, Morse DL. A nosocomial outbreak ofmalaria associated with contaminated catheters and contrast medium ofa computed tomographic scanner. Infect Control Hosp Epidemiol1999;20:22-25.

5. Abulrahi HA, Bohlega EA, Fontaine RE, al-Seghayer SM, al-Ruwais AA.Plasmodium falciparum malaria transmitted in hospital through heparinlocks. Lancet 1997;349:23-25.

6. Dziubek Z, Kajfasz P, Basiak W. Hospital infection of tropical malaria inPoland. Wiadomosci Lekarskie 1993;46:860-863.

7. Al-Saigul AM, Fontaine RE, Haddad Q. Nosocomial malaria from conta-mination of a multidose heparin container with blood. Infect ControlHosp Epidemiol 2000;21:329-330.

8. Sabatinelli G, Majori G. Malaria surveillance in Italy: 1986-1996 analysisand 1997 provisional data. Eurosurveillance 1998;3:38-40.

9. Romi R, Boccolini D, Majori G. Malaria surveillance in Italy: 1997 analy-sis and 1998 provisional data. Eurosurveillance 1999;4:85-87.

10. Baldari M, Tamburro A, Sabatinelli G, et al. Introduced malaria inMaremma, Italy, decades after eradication. Lancet 1998;351:1246-1248.

11. Wooden J, Kyes S, Sibley CH. PCR and strain identification inPlasmodium falciparum. Parasitology Today 1993;9:303-305.

12. Menegon M, Severini C, Sannella A, et al. Genotyping of Plasmodium fal-ciparum gametocytes by reverse transcriptase polymerase chain reac-tion. Mol Biochem Parasitol 2000;111:153-161.

13. Viriyakosol S, Siripoon N, Petcharapirat C, et al. Genotyping ofPlasmodium falciparum isolates by the polymerase chain reaction andpotential uses in epidemiological studies. Bull World Health Organ1995;73:85-95.

14. Snounou G, Beck HP. The use of PCR genotyping in the assessment ofrecrudescence or reinfection after antimalarial treatment. ParasitologyToday 1998;14:462-467.

15. Ohrt C, Mirabelli-Primdahl L, Karnasutra C, Chantakulu S, Kain KC.Distinguishing Plasmodium falciparum treatment failures from reinfec-tions by restriction fragment length polymorphism and polymerasechain reaction genotyping. Am J Trop Med Hyg 1997;57:430-437.

16. Center for Disease Control. Nosocomial transmission of hepatitis Bvirus associated with a spring-loaded finger stick device: California.MMWR 1990;39:610-613.

17. Polish LB, Shapiro CN, Bauer F, et al. Nosocomial transmission ofhepatitis B virus associated with the use of a spring-loaded finger-stickdevice. N Engl J Med 1992;326:721-725.

18. Quale JM, Landman D, Wallace B, Atwood E, Ditore V, Fruchter G. Dejavu: nosocomial hepatitis B virus transmission and fingerstick monitor-ing. Am J Med 1998;105:296-301.

19. Shaughnessy AF, Slawson DC. Pharmaceutical representatives. BMJ1996;312:1494.

This content downloaded from 195.78.109.14 on Wed, 14 May 2014 12:42:47 PMAll use subject to JSTOR Terms and Conditions