Maltese Fever

MALTESE FEVER is known under various names, of which the most common is undulant, or wave-like fever (febris undulans, French fievre ondulante, English undulant fever). Other names exist: Mediterranean fever, melitococci, and septicaemia melitensis. Numerous outdated designations for the disease are numerous: Gibraltar fever, Cyprus fever, Cretan fever, Neapolitan fever, goat fever, fievre folle, febricula typhosa, febris sudoralis (miliaria), phthisis mediterranea, typhoidea atypica, typho-malarial fever, and others). None of the names, however, is satisfactory. It would perhaps be most correct to designate this disease as 'brucellosis,' as it actually is. In light of modern data, M.F. can be characterized as an infectious disease of humans, caused by pathogenic varieties of bacteria belonging to the genus Brucella. From an etiological standpoint, it represents a particular case of brucellosis. Brucella infection is widely prevalent in various domestic animals (cattle and small ruminants, pigs). Sick animals form a reservoir of the virus, from which human infection occurs. From a clinical standpoint, M.F. is characterized as a relapsing, usually febrile disease of indeterminate duration, proceeding with considerable diversity of accompanying symptoms and complications. The disease may be accompanied by a wave-like type of fever curve, from which its name 'undulant fever' is derived. History. As an independent clinical entity, M.F. was differentiated in 1863 by Marston on the island of Malta (hence the name M.F.). There in 1887, Bruce discovered the causative agent of the disease, named Micrococcus melitensis. In 1897, Wright found that serum from M.F. patients specifically agglutinates cultures of the causative agent; Wright's findings formed the basis for the widely accepted serodiagnosis of M.F. using the agglutination reaction (Wright reaction). During 1904-06, an English commission working on the island of Malta established that the epidemiology of M.F. is linked to goats; infected with the micrococcus under natural conditions, goats excrete the causative agent in urine and milk and are thus a source of infection for humans (infection through milk). Further observations revealed that M.F. is widely distributed in hot countries of both the Old and New Worlds. Endemic foci of the disease generally coincide with corresponding enzootics among small ruminants (goats and partly sheep). In 1912, Negre and Raynaud, among cultures isolated from M.F. patients, differentiated an original variety, naming it Micrococcus paramelitensis. Thus the possibility of multiple etiology of M.F. due to related variants of the causative agent emerged. In 1918, Evans discovered the extraordinary similarity of the Maltese micrococcus to the causative agent of infectious abortion in cattle (Bact. abortus), discovered as early as 1896 by Bang (see Infectious Abortion). This gave Meyer and colleagues grounds to unite the causative agents into one group with the generic name Brucella (Br. abortus, Br. melitensis). Subsequently it was found that a disease clinically indistinguishable from M.F. can undoubtedly be caused by Bang's microbe - Brucella abortus (the first culture of Br. abortus from a sick person was obtained in 1924 by Keefer in Boston). Later observations showed that M.F. caused by Bang's microbe (cattle and partly pigs) is found everywhere, and in some northern regions it exists endemically and is very widespread (Kristensen in Denmark during 1927-29 established 500 cases). Thus the concept of M.F. as an exotic southern infection caused by the Maltese micrococcus gradually gave way to the modern view of this disease as a brucella infection having distribution in both southern and northern latitudes. This circumstance in turn placed the problem of Maltese fever in an entirely new plane. The etiology of Maltese fever is related to pathogenic variants of bacteria of the genus Brucella. The main subgroups of this genus are varieties of the type Br. melitensis (parasite of small ruminants and humans) and type Br. abortus (parasite of cattle and pigs). Both subgroups do not differ from each other either morphologically or biologically. These are the smallest bacteria of ovoid (0.3-0.4 μ) or more elongated (0.8-1.8 μ) form. They are immobile, do not form spores, are easily stained by aniline dyes, are Gram-negative; they grow well but slowly on ordinary nutrient media of neutral or slightly alkaline reaction, are indifferent to carbohydrates, do not coagulate milk, do not liquefy gelatin; in relation to oxygen they are facultative anaerobes, but Br. abortus cultures in the first generations usually succeed better under anaerobic conditions or in an atmosphere of CO₂. It is further characteristic of both subgroups the formation on agar of homogeneous or finely granular colonies (Zdrodovsky and Voskresensky), the formation of a stable emulsion that does not change when heated to 90° (negative thermoprecipitation Burnet), and diffuse growth in broth. Cultures possess excellent agglutinogenic and agglutinable properties, why the infection with them is easily diagnosed using the agglutination reaction. Finally, cultures are highly pathogenic for guinea pigs by the most diverse methods of infection (cutaneous, conjunctival, oral and parenteral). Br. melitensis and Br. abortus are not differentiated by direct agglutination and only inconsistently differ by the method of absorption of agglutinins (Zdrodovsky and Brenn). The second subgroup of the Brucella species consists of variants of the type Br. paramelitensis (goats and humans) and Br. paraabortus (cattle). From the previous type variants, the latter differ in the formation of coarse granular colonies (Zdrodovsky and Voskresensky), a tendency to spontaneous agglutination in salt emulsions, giving positive thermoprecipitation by Burnet, growth in the form of flakes in broth with subsequent sediment. Furthermore, cultures show very low agglutinogenic capacity, why in infections with Br. paramelitensis diagnosis by agglutination often fails. By direct agglutination and especially by the method of absorption, para-variants are very well differentiated from Br. melitensis and Br. abortus. According to observations by Zdrodovsky and Voskresensky, the para-variants under consideration, when they have coarse granular colonies, are very weakly virulent or completely non-virulent for guinea pigs. Obviously Br. paramelitensis and paraabortus represent labile varieties, particularly prone to dissociation with transition in the final stage to the avirulent 'R' type (Zdrodovsky). According to Burnet, Br. melitensis cultures can be artificially converted into Br. paramelitensis cultures, which speaks for the conditional nature of the latter and confirms their origin through dissociation. The same applies to Br. abortus cultures. Attempts have been made to classify the Brucella group based on the method of absorption of agglutinins; Evans differentiated five main subgroups (Br. melitensis I to V, Br. paramelitensis, Br. abortus and paraabortus) and 3 less defined variants. However, the serological classification is very relative and has only formal significance. In particular, variants of Br. abortus and melitensis are pathogenic for humans regardless of their serological position in the Brucella group. For example, Azerbaijani and partly Tunisian cultures of Br. melitensis are very close or almost identical with Br. abortus. On the other hand, some cultures of Br. abortus from northern latitudes (e.g. from the Vitebsk region) are indistinguishable by the method of absorption from type cultures of Br. melitensis (Zdrodovsky, Brenn and Voskresensky). Therefore it is expedient to divide the variants according to their origin only into two main subgroups: Br. melitensis (brucellosis of small ruminants; obligate pathogenicity for humans) and Br. abortus (brucellosis of cattle and pigs, facultative pathogenicity for humans). Along with this, the isolation of Br. paramelitensis (and paraabortus) as dissociating variants is justified. The epidemiology of M.F. is related to brucellosis in animals. The decisive importance is brucellosis of small ruminants and especially goats, since their causative agents are obligately pathogenic for humans. Brucellosis in goats (as well as in rams) often proceeds without any symptoms and is recognized only by laboratory methods (agglutination, cultures). At the same time, animals obviously are not cured (Zammit). The causative agent is excreted in urine and especially in milk (selective localization in lymph glands in the udder area - Burnet). Raw milk of these animals and products made from it (butter, cheese, etc.) are the main source of infection for humans, and alimentary infection has particular importance in general. It is important to note that the causative agent is resistant and can persist for a long time in the external environment.

Approximate survival of the pathogen on various substrates: milk-up to 20-60 days, sour dairy products and cheese-up to 21 days; unsalted cheese from goat's milk-up to 14-44 days and more, butter-up to 25 days, fruits-up to 22 days, water-up to 10-71 days, urine of patients-up to 20 days, dust infected with urine-up to 30 days, dry soil-up to 42 days, moist soil-up to 72 days (Dopter and de Lavergne, Lustig and Vernoni).-The non-alimentary method of infection occupies second place. This includes infection through damaged skin (in milkers), through the mucous membranes of the mouth and nose, through the respiratory tract, through the conjunctiva, and finally through the urinary system (sexual infection). If for alimentary infection milk and dairy products are particularly important, for the latter type of infection urine is of primary importance.-Brucellae of cattle (and pigs) in different places apparently have unequal epidemiological significance. In some places these brucelloses are only episodically accompanied by human diseases; conversely, in other regions (e.g. Denmark) they are the source of fairly widespread human infection. The reasons for the unequal pathogenicity of Brucella abortus variants for humans remain little clarified. It should be noted that in some places the epidemiology of Maltese fever was associated with the corresponding infection of other animals, in particular rabbits and dogs. In some foci of Maltese fever, brucellar infection of horses was found (e.g. in the Salsky district--Roznatovsky, Minervin). The possibility of spontaneous infection in cats is not excluded. Pigs very easily contract brucellosis and in laboratory conditions can be a source of infection for personnel (in the Azerbaijan Institute, as a result of the accidental introduction of infection into the vivarium, an epizootic developed that affected about 400 pigs; as a result, 10 personnel members were infected). Distribution. Maltese fever as a human disease, epidemiologically associated with brucellosis of small ruminants (goats and partly sheep), forms endemic foci mainly in warm and hot countries. Its geographical distribution is as follows: Europe-Spain, Portugal, southern France, Italy, Greece, Turkey, Mediterranean islands (Malta, Corsica, Sardinia, Sicily, Balearic Islands, Crete, Cyprus); Asia-Asia Minor, Syria, Palestine, Arabia, Persia, British India, Java Island, Philippine Islands, etc.; Africa-North Africa (Morocco, Algeria, Tunisia, Tripolitania, Egypt), tropical Africa (eastern, western, central), South Africa (Cape colonies, Transvaal), Canary Islands; America-North America (southern states, Mexico), South America (Brazil, Uruguay, Venezuela, Peru), West Indies; Oceania-Fiji Islands.-As a disease associated with brucellosis of large cattle (and partly pigs), Maltese fever has recently been discovered in the following countries: Europe-Denmark, Holland, Sweden, England, France, Germany, Switzerland, Austria. Italy, Poland; America-Canada, USA. It should be noted the foci of Maltese fever in South Africa (Rhodesia). On the territory of the USSR, Maltese fever was almost unknown until recently (1 case was discovered in 1911 in Moscow by Marcynowski and 2 cases in 1911 by Kramnik in Ashgabat).-The history of this disease in the USSR actually begins in 1922, when Kryukov and Smirnov discovered 5 cases of Maltese fever in Tashkent and at the same time Zdrodovsky established 6 cases in Azerbaijan, and the latter author for the first time proved the endemicity of this infection in the USSR (findings of human diseases and homogeneous infection in goats). Subsequently, findings of Maltese fever were widely confirmed, and the following picture of the endemic spread of this disease on the southern outskirts of the USSR was clarified: a) Caucasian district: 1) Transcaucasia-Azerbaijan, Armenia, Georgia (Zdrodovsky and colleagues, Popov, Ieaakyan). 2) Caspian strip of the Caucasus (Zdrodovsky and colleagues). 3) Black Sea strip of the Caucasus (Lindtrop). 4) North Caucasus region (Zavadsky, Roznatovsky, Starkova, Ritenko-Zavarzina, Minervin). b) Transcaspian district: 1) Uzbekistan (Kryukov and Smirnov). 2) Turkmenistan (Kramnik, Yefremov, Stepanov). 3) Tajikistan (Rosen and colleagues). Undoubtedly, Maltese fever has a wider distribution, but is not recognized by physicians, being generally little known. In particular, its distribution along the entire coast of the Caspian and Black Seas is reliably established; probably in the southern Volga region and southern districts of Ukraine.-The question of Maltese fever caused by Brucella abortus has barely begun to be studied in the USSR. But already the first attempts to search for this disease have given positive results in other places of the Union. Thus, individual cases of the disease have been found in the regions of Minsk, Leningrad, Moscow, Vitebsk, Novocherkassk and the Salsky district. Undoubtedly the disease is widespread throughout the Union, but is not recognized.--Due to the novelty of the issue and the unknown nature of the disease itself among Russian physicians, statistics on Maltese fever in the USSR are almost absent and are limited to episodic data. In total, from 1922 to 1929, about 500 cases were registered, including about 200 cases in Azerbaijan. More indicative are the results of mass surveys. For example, in Azerbaijan, when examining 292 people, the infection was found in 4.6%, in the Salsky district of the North Caucasus region, when examining 806 residents from 46 villages, the infection was found in 238 people, in the Ashgabad district of Turkmenistan, when examining 828 people, infection was established in 34%. Thus, at least in some districts of the USSR, Maltese fever is very widespread. Pathological anatomy of Maltese fever in humans is little studied. The most characteristic is the lesion of the lymphohematopoietic system (spleen, bone marrow, lymph glands). Against the general background of sharp hyperemia and hemorrhages (e.g. in the intestine), regenerative changes of specific elements (degeneration and necroses) prevail, and only partly reactive processes (infiltration). Degeneration of parenchymal organs (heart, kidneys) is absent or weakly expressed, which obviously explains the relative benignity of the disease. In a number of cases of abortive etiology, cirrhotic changes of the liver, ascites, enlargement of the spleen are noted, i.e. combinations of the disease with the hepatolienal syndrome. Information on pathological anatomy is, however, insufficient and should be supplemented by corresponding observations on pigs, which are a classical object for studying brucellar infection. According to numerous observations of the Azerbaijan Institute (Kolesnikov), the pathological anatomy of Maltese fever in pigs may not be the same depending on the virulence of the dose of the pathogen and the resistance of the organism. Two types can be distinguished-areactive and reactive. In the first case, the infection borders on symbiosis and is accompanied by almost no changes except for mild hyperplasia of the lymphatic apparatus. In the second case, the changes, on the contrary, are sharp, specific, and of a tuberculous nature. The basis of the changes lies in the mobilization of mesenchymal elements and especially of reticulo-endothelial cells. As a result, diffuse cell proliferations are formed or tuberculous-type granulomas consisting of epithelioid cells with an admixture of giant cells. Tubercles can have different localizations (spleen, lymph glands, liver, lungs, bone-joint system, testicle, etc.). The fate of the proliferations is twofold-sclerotic transformation or suppuration. The latter are particularly prone to lesions of the bone system and testicle. In all its properties, Maltese fever resembles tuberculosis. This similarity concerns the general course of the disease (often clinically confused with tuberculosis), pathological anatomy, immunity, and allergic relations. The latter are very close to tuberculous. As shown by Burne, the broth filtrate of the culture-'melitin'-reproduces the reactions in Maltese fever that tuberculin gives in tuberculosis. By analogy with tuberculosis, brucellar infection 'premunizes' and gives resistance to superinfection, and in the latter case the body responds with allergic reactions similar to tuberculosis. The clinical picture of Maltese fever is very diverse. In a number of cases, the infection can proceed without symptoms or with very minor clinical manifestations (general malaise, short-term temperature increases to 37-38°, headaches, mild neuralgias and rheumatic pains, etc.). These asymptomatic and ambulatory forms of Maltese fever are quite common in endemic foci, as was shown by Shaw for the island of Malta and confirmed for Azerbaijan, in particular on a large material (15) with laboratory infection. It should be borne in mind, however, that ambulatory forms, even after a long course, can resolve into very severe attacks of the disease with various complications. To these cases of late exacerbation of the disease are closely related cases with 'primary prolonged latency' in the proper sense, when the infection for a long time (according to observations by Zdrodovsky and colleagues for more than 7 months) can remain without any manifestations, but later gives a bright clinical picture.-In the usual course, the incubation period in Maltese fever averages from 1 to 2-3 weeks.

The most characteristic feature of the infection is considered to be fever. However, the forms of febrile reaction are very diverse and can only be conditionally divided into the following types: 1) irregular recurrent fever (forma indeterminata et lenta of Italian authors)—a very common type; 2) undulating fever—the most characteristic type of M. f.; 3) intermittent fever; 4) continuous fever (febris continua)—the rarest types; 5) malignant typhoid-like fever—a type characteristic of the most severe, often fatal forms. Particularly distinctive is the undulating type of temperature (see figure), in which febrile periods alternate wave-like with afebrile intervals. The duration of febrile waves varies from 1 to 6 weeks (on average 2-3 weeks), with afebrile intervals from 4-7 to 10-14 days (sometimes several weeks). The number of febrile waves varies from 2-3 to 6-7 or more with a total duration of fever from several weeks to many months. The other symptoms and complications observed in M. f. are extremely diverse. From the skin side, a special tendency to sweating (f. sudoralis) is noted; various rashes (roseolar, papular, pustular, scarlet fever-like erythema, etc.) are also observed, as well as petechiae and hemorrhages; desquamation of the epidermis, hair loss, and changes in nails (brittleness) may occur. Sometimes the skin manifestations are so pronounced that one speaks of the exanthematous form of the disease (Grocco). Lymph glands may enlarge, and sometimes recurrent polyadenitis is observed (Budylin and others). The spleen is usually enlarged; sometimes the liver is enlarged (jaundice may occur). From the intestine side, constipation is common. For the cardiovascular system, irregular heart activity, rapid pulse, and decreased blood pressure are characteristic; rarely endo- and pericarditis, phlebitis, and thrombophlebitis are observed.--Blood picture: leukopenia and monocytosis; in focal lesions, leukocytosis may occur; in exhausting course—secondary anemia. From the lungs side, in addition to bronchitis, recurrent bronchopneumonia, as well as peculiar alveolitis with bloody sputum and hemorrhages (simulation of tbc) may be observed. Lesions of the nervous system are very characteristic: headaches, insomnia, irritability, and general depression—usual companions of the disease; neuralgias with various localization, often very painful, are very frequent; the latter may be replaced by persistently recurrent polyneuritis; radiculitis, as well as muscle paresis and paralysis, are also encountered; sometimes meningitis may develop (simulation of tbc and syphilis). The synovial system is involved in the process very often; very painful tendovaginitis (simulation of gonorrheal metastases) and recurrent and migratory bursitis are not uncommon; serous effusions into the joints may be observed. The bone system can also be affected with the formation of periostitis and osteitis, sometimes with subsequent suppuration; coxitis and spondylitis simulating tbc deserve special attention. Among the lesions of glandular organs, epididymitis and orchitis (up to 6%), parotitis, and sometimes mastitis in women must be mentioned. Finally, complications from the eyes (congestive papilla, amblyopia, and atrophy of the optic nerve) have been described.-The mortality rate in M. f. is low; according to Chelli (SeH), it ranges from 1.4% to 8.9%. Epidemics on the continent often run more severely. The duration of the disease is indefinite. According to Bassett-Smith, for 552 cases the average duration was 4 months. But the disease can drag on for years (in one of Zdrodovsky's cases the duration was more than 4 years). Even very long periods of apparent recovery cannot guarantee against subsequent relapses (secondary long latency). The diagnosis of M. f. can be made clinically with greater or lesser probability only in particularly typical cases (undulating fever). In practice, the disease is often not recognized by clinicians and is diagnosed under various names (malaria, typhoid fever, paratyphoid, tbc, sepsis, etc.). Laboratory research methods are of decisive importance for the diagnosis of M. f. The most accurate diagnostic method is bacteriological examination of the patient with isolation of the pathogen's culture. The latter is localized most consistently in the spleen, lymph glands, and bone marrow; in humans, it is often also found in the blood. The microbe is excreted with urine, milk, and in the presence of a pulmonary process—with sputum. Finally, in cases with local complications (joint lesions, bone lesions, orchitis, abscesses, etc.), the microbe is found in the corresponding exudates. For diagnostic purposes, blood culture on broth is usually used (up to 10 cm³ of blood in several flasks with 50-100 cm³ of ordinary broth). The method gives a positive result in 68-85%. Blood culture is possible already from the first days of the disease, but can also be obtained in very late periods (almost after a year in Eyre's case). Cultures are preferably taken at the height of fever, but cultures are also obtained in subfebrile condition and even at normal temperature (observations of Zdrodovsky). It is important to keep in mind that the pathogen can develop very slowly, and therefore cultures should be kept in an incubator for at least 2 weeks. The Wright reaction (agglutination) is especially widely used for the diagnosis of M. f., which is quite specific, usually very well expressed, obtained consistently (exceptions—see below), and can be observed already from the first days of the disease [in one of Zdrodovsky's cases the Wright reaction (1:250) was obtained on the first febrile day]; the reaction is quite convincing if it is macroscopically well expressed with a serum dilution of 1:100; usually it is obtained at higher dilutions (1:200-1:1000 and higher), and in individual cases the titer reaches 1:10000 and higher (up to 1:50000, according to Eyre). The Wright reaction is set with a live culture (dangerous!) or with a killed emulsion; a ready diagnosticum can be used successfully. Instead of Br. melitensis, a selected dilution of Br. abortus can be used. The disadvantage of the Wright reaction is that it is usually not obtained in Br. paramelitensis infection. Thus, the absence of this reaction by no means excludes brucellosis. Recently, Burne proposed a very simple yet reliable diagnostic test with melitin (filtrate of Br. melitensis broth culture). The preparation in an amount of about 0.1 cm³ is injected intracutaneously. In case of brucellosis, a painful swelling and redness form at the injection site already after 6-8 hours. The reaction is quite specific, is detected already from the 8th day of the disease, and is constant. According to Burne, the melitin test also succeeds in cases where blood culture or agglutination reaction is negative (especially in Br. paramelitensis infection). Thus, Burne's test is the most sensitive. Mass observations by Zdrodovsky using it on guinea pigs fully justify this. Instead of melitin, abortin (filtrate from Br. abortus culture) can also be used successfully. Incidentally, Burne's test works well in humans and guinea pigs, but does not work in goats. The treatment of M. f. is mainly symptomatic. Attempts at specific therapy can be briefly summarized as follows: vaccination (and autovaccination) has been and is used by many authors, but its results are contradictory. It is possible that the contradictory nature of observations is explained by the diversity of the vaccination technique itself (dosage, selection of cultures). In general, the question of vaccinotherapy requires further development. Serotherapy of M. f. is even less developed, although attempts at it have been made repeatedly by various authors. Recently, the research of Sergent and Lheritier seems to have outlined rather optimistic prospects in this direction. However, control observations of the Azerbaijan Institute did not confirm the data of the French authors; it turned out that sera from hyperimmunized horses at best have only an analgesic effect (in the presence of algias and others). Apparently, more favorable effect is given by convalescent serum (sometimes a curative effect). According to observations of some authors, autohemotherapy has a favorable effect on M. f. The fight against M. f. is very difficult, since the disease is epidemiologically linked to animal brucellosis. For personal and public prevention, the following measures are recommended: 1) correct diagnosis and registration of diseases; 2) disinfection of patient's excretions (urine) and all objects suspected of being contaminated (linen); 3) exclusion of raw milk and raw dairy products from consumption; 4) the same in relation to products that may be contaminated with urine (fruits, greens, etc.); 5) observance of precautionary measures by persons caring for patients and coming into contact with animals; 6) sanitary-veterinary supervision of dairy farms; 7) prohibition of making cheese, butter, etc. from unpasteurized milk; 8) sanitary education.

Preventive measures: 1) systematic examination of animals using the Wright reaction to identify sick individuals; 2) destruction of infected animals; 3) inspection of imported animals from endemic foci. For the prevention of the disease in humans and animals, some authors recommend periodic vaccinations, although this method is generally little developed. Special caution is required from persons working with Brucella melitensis cultures, as the causative agent is extremely contagious (intralaboratory infections are observed very often).