Atypical Pneumonia and Q Fever in Second World War

Q fever was one of the two new diseases of armies that came to be recognised during the Second World War. In Italy in the winter of 1943–44 there were several epidemics of what was called primary atypical pneumonia, which are now thought to have been Q fever. This was when New Zealanders first experienced it. page 583 In February 1944 there were between 70 and 80 cases admitted to 2 NZ General Hospital at Caserta, 48 of them arising from 7 NZ Anti-Tank Regiment, and sporadic cases from other divisional units then bivouacked 10 miles south of Cassino. Some 30 cases were evacuated to hospital from the Anti-Tank Regiment within two or three days. Most of the cases began in one battery and nearly all came from two batteries. The disease was supposed to be spread by droplet infection, but in an investigation the Consultant Physician 2 NZEF and the RMO of the unit found little to support this contention. The Consultant Physician was inclined to the view that the disease was a virus infection.

The area occupied by the unit was a morass of mud as the result of constant rain, and the men were living in square tents holding four to six men, or else in small bivouac tents. Battery Headquarters was in some Italian farm buildings, but the men had not been living in Italian houses in any numbers and their general health was first class. Cooking was by individual battery arrangements. Sanitary facilities in the area were primitive and unpleasant.

Clinically the sudden onset with severe frontal headache and prostration were the salient features of the outbreak. Another striking feature was that the early stages were characterised by a complete absence of catarrhal symptoms. There was no coryza or sinusitis, and cough was unusual in the first day or two. The cough which developed about the third day never became troublesome or painful; there was little if any sputum, and what there was was clear and mucoid. The patients for the most part were very toxic, but physical signs were scanty throughout the illness. Treatment was only symptomatic, and there was no response to sulphonamide therapy. X-ray examinations, however, revealed typical shadows which were found to persist for some weeks after apparent clinical recovery. The consolidation might be generalised, or merely at the apex or in the region of the interlobar fissure. Often the fan-shaped appearance of the consolidation was characteristic. Increased bronchial markings were also a feature. There was cellular infiltration around the alveoli and a certain amount of intra-alveolar exudation and exfoliation of epithelium.

Turner reported a similar epidemic of 280 cases from the Naples area. During the same year the Germans, who were still in occupation of Greece, had their troubles with primary atypical pneumonia, which was referred to by them as the ‘Balkan Grippe’. Dr Caminopetros of the Pasteur Institute of Greece was able to establish the infective agent in guinea pigs by injection of blood from patients in the acute stage of the illness. Subsequently blood page 584 from these guinea pigs was flown to the United States and investigated at the laboratories of the Commission on Acute Respiratory Diseases at Fort Bragg, North Carolina, and the Rickettsia burneti was isolated.

Epidemics of primary atypical pneumonia recurred in Italy in the winter of 1945 among British and United States troops. Adams, Staveley, Rolleston, Henley and Caughey recorded a study of fifty cases which occurred among New Zealand and British troops in the Caserta and Naples area. Four epidemics of a similar nature, among United States troops in northern Italy, were investigated and the R. burneti was isolated as the infective agent, and by serological tests it was established that at least three-quarters of the cases of atypical pneumonia occurring in the same area were, in fact, cases of Q fever. Details of these epidemics have been published in the American Journal of Hygiene (1946).

In May 1945 an epidemic of primary atypical pneumonia occurred in mid-Atlantic among a group of airmen who had recently been stationed at Grottaglie air base in southern Italy in the Taranto area. These were fully investigated when the airmen arrived at Camp Patrick Henry, Virginia. Epidemiological and etiological studies were made by the Commission on Acute Respiratory Diseases and it was demonstrated that the infective agent was the ‘Balkan Grippe’ type of the R. burneti.

Clinical Features

Clinical details of the epidemic which was investigated at 2 NZ General Hospital have been published in the British Medical Journal, 16 February 1946, Vol. 1, page 227. The principal clinical features will be here summarised.

The epidemic occurred between February and April 1945, and during this period 511 cases were reported in the area, 161 of which were treated at 2 NZ General Hospital. Fifty consecutive cases were studied in detail clinically and by serial pathological and radiological investigation. In most there was a prodromal period of about six days. The actual onset was abrupt in 96 per cent of cases. Severe headache, malaise, lassitude, and anorexia were the most constant symptoms. Pyrexia from the onset averaged 8.6 days: was over 103 degrees in 70 per cent of cases and defervescence was by lysis in 86 per cent. The pulse followed the temperature though showing a tendency to a relative bradycardia. The respiratory rate was little affected. Cough occurred about the fifth day and was present in 94 per cent, although it was not an outstanding feature. Sputum was scanty and in 28 per cent contained blood. Chest pains occurred in 46 page 585 per cent of cases. (These symptoms were similar to those of the patients treated in 2 NZ General Hospital a year previously.)

Severe toxaemia was a feature of some and one-third showed generalised rhonchi on admission, but the most characteristic sign was a localised patch of sticky persistent crepitations heard on an average on the sixth day from the onset of the acute symptoms. Pleural friction occurred in 26 per cent of those with chest pains. There was usually enlargement of glands and the spleen was palpable in 36 per cent of cases. Scanty small pink macules fading on pressure were observed on the chest, back, and flanks in the early stages of the disease in 34 per cent of cases.

The results of serial pathological investigation can be summarised in brief. The white count revealed a slight polymorphonuclear response, followed by a slight depression which was maximal at the end of the first week. Thereafter there was a rise of polymorphs and lymphocytes, reaching a peak at the sixteenth to eighteenth day. Differential counts showed a slight relative lymphocytosis after the initial period. The blood sedimentation rate was elevated for two to three weeks. Cephalin-cholesterol flocculation was insignificant early in the disease, but increased rapidly after convalescence had been established in the second week, thereafter falling slowly. Using horse cells, a significant positive heterophil antibody reaction was found in 36 per cent of the cases at some stage of the disease. All but three, which were weakly positive, gave negative tests for cold agglutinins at all stages of the disease.

Posterior-anterior and lateral X-ray studies were made in all cases. The characteristic findings were the localisation of the lesion to one or more broncho-pulmonary segments. The infiltrations could be described as hazy, mottled densities. These investigations also revealed the importance of the lateral studies for the demonstrations of lesions situated behind the heart shadow or in that portion of the lung situated behind the summit of the dome of the diaphragm. In the majority complete radiological resolution occurred within six weeks of the onset. No specific treatment was found to be of any use and sulphonamides had no effect.

Serological investigation of some of these cases was carried out two years later by Caughey and Dudgeon. Sera from twenty of the cases were tested for complement fixation against antigen prepared from a strain of the Rickettsia burneti isolated in Italy. Nineteen gave a positive complement fixation test. The sera were also tested for antibody to the virus of psittacosis, and all were negative. In view of these findings, it seemed reasonable to assume that the infective agent in these cases was the Italian strain of the Rickettsia burneti.

page 586

Epidemiological Studies

1.	In Australia epidemiologcal studies revealed a widespread incidence of the Rickettsia burneti. The native rodent, the bandicoot, was found to be susceptible, and evidence of latent infection was found in other bush animals, in water rats, other native rats, and also in cows. The Rickettsia burneti was isolated from ticks (haemaphysalis humerosa) collected from the bandicoots, and the faeces of these infected ticks were found to be highly infectious and capable of infecting guinea pigs when applied to the skin. In Derrick's opinion, this tick, the H. humerosa, is the vector among bandicoots. This tick was readily infected with the R. burneti by feeding on infected guinea pigs. The infection was transmitted from larvae to nymph, and from nymph to adult. The R. burneti were found to be confined to the lumen of the gut and faeces were heavily infected. Three other ticks could also be infected— rhipecephalus sanguinius, ixodes holocyclus, and haemaphysalis bispinosa—by feeding on infected guinea pigs, and could transmit infection to their host. Each, therefore, is a potential vector, and Derrick's conclusion is that the haemaphysalis bispinosa is the probable source of the human infection in Queensland and that the mode of infection is by inhalation of infected faecal dust from these ticks.
2.	In Italy the possible route and sources of infection were examined in detail in some of the northern Italian epidemics, and although no vectors, such as ticks or fleas, were incriminated, it was found that the disease often made its appearance in units occupying farm billets and where men were living in close proximity to animals such as cattle, rats, and also pigeons. Various forms of mites were discovered in several of these billets, but it was thought that cases probably occurred by inhalation infection arising from the dust and droppings. The complete absence of any insect bites in any of the cases, and also the uniform picture of pulmonary involvement, suggested that the route of infection was via the upper respiratory tract, such as probably occurred in the laboratory infection with the same agent (Robbins and Rustigan, 1946), and which is also seen in at least laboratory infections with other rickettsial diseases such as epidemic murine and scrub typhus. It was also found that many of the civilian population in the area where these cases occurred showed a high level of antibody to Q fever, suggesting that the infection was endemic in the local population.
3.	In the New Zealand hospital cases most of the patients came from a British infantry training depot where men were living in hutments and where there was a rapid turnover of personnel. The page 587 staff of the hospital escaped lightly. A strict isolation technique was carried out in the wards until the temperature had been normal for a week, or until productive cough had ceased. The regimental aid post orderly, who treated men daily from the infantry training depot, and the hospital librarian who visited the wards, both contracted the disease, but no case occurred among nurses who were nursing the cases.

References

E. H. Derrick	Q Fever, New Fever Entity: Clinical Features, Diagnosis and Laboratory Investigation. Medical Journal of Australia, 2, 281–99.
	Epidemiology of Q Fever, Journal of Hygiene, 1944 (43,357).
F. M. Burnet and M. Freeman	Experimental Studies on Virus of Q Fever. Medical Journal of Australia, 1937 (2, 299–305).
R. W. D. Turner	Lancet, 1945 (1,493). The Commission on Acute Respiratory Diseases. Identification and Characteristics of the Balkan Grippe Strain of Rickettsia burneti. American Journal of Hygiene, Vol. 44, No. 1, July 1946.
A. B. Adams, J. M. Staveley, G. L. Rolleston, W. E. Henley, J. E. Caughey	British Medical Journal, 1946 (1, 227).
F. C. Robbins, R. Rustigan	American Journal of Hygiene, 1946 (44).
M. Feinstein, R. Yesner, J. L. Marks	American Journal of Hygiene, Vol. 44, No. 1. 72–87, July 1946.
J. E. Caughey and A. Dudgeon	British Medical Journal, 1947 (Vol. ii, 684).