ORIGINAL ARTICLES The Pathogenesis and Radiological Features of Hut Lung Authors Eric D Bateman MBChB, MD, FRCP, OCH Associate Professor and Head Respiratory Clinic Groote Schuur Hospital and University of Cape Town David Stickells MBChB, OP aBST (SA), FCP(SA) Registrar Hillel Goodman FRCR, FFRad(O){SA) Principal Specialist Radiologist I n 1967 the term "Transkeisilicosis"was proposed by Palmer and Daynes for a form of domestically-acquired pneumoconiosis found in women from the rural Transkei and Ciskei which radiologically was indistin- guishable from silicosis seen in South Afri- can gold miners.I The source of silica was thought to be from quartz-containing grind- ing rocks used for preparing the daily stand- ard meal of ground maize. Post-mortem histology on two cases confirmed the ap- pearance of silicosis, and a small epidemiologic study of healthy young nurses in the Transkei confirmed early changes of pneumoconiosis in approxi- mately 7%.1 Subsequently, on the basis of findings in a larger and more detailed case collection conducted in our clinic involv- ing the study of clinical presentation, lung physiology, bronchoalveolar lavage cell counts and histological appearance of lung biopsies, we concluded that although sili- cosis may account for some of the changes in some patients with this clinical entity, it appears that inhalation of non-quartz con- taining dust and smoke from bio-mass fuelled fires might be of greater significance in the aetiology/ The more general term "hut lung" was therefore suggested. Rec- ognition of this condition is clearly impor- 4 SAJOURNAL OF RADIOLOGY· May 1996 tant in distinguishing it from other forms of interstitial lung disease,most notably dis- seminated or miliary tuberculosis, and sarcoidosis,and from a public health stand- point, it is a preventable disease leading to significant complications and mortality. This paper will highlight aspects of the pathogenesis and radiologicalfeatures of this condition. Pathogenesis The concept of dust diseases (pneu- moconiosis) occurring in settings other than the confines of the workplace is being increasingly recognised. Silicaparticles may be found'in the lungs of persons living in deserts,' and have even been found in Egyp- tian mummies" Although silicosis is rare in this setting, cases of severe silicosis and even progressive massive fibrosis have been found in inhabitants of Himalayan villages exposed to dust storms.' Outdoor expo- sure to dusty environments such as tractor ploughing in the Free State" and farming activities in rural Canada have also been shown to cause disease.' Silica is the best .recognised fibrogenic dust in these studies, but several lines of evidence suggest that silica alone may not be responsible for the nodules seen in hut lung. 1. Although respirable free silica can be identified in the ground maize, estimates of silica particles in the lung are low, even within nodules? 2. A small field study in which atmospheric silica was sampled by means of personal samplers on subjects grinding with sand- stone (100% quartz), confirmed that the concentration of respirable quartz in sur- rounding air was low, and that the cumula- tive time-weighted exposure was well be- low that recommended for industry and mines. Such levels would not be ex- pected to cause pneumoconiosis, and certainly not life-threatening disease.' 3. Similar radiological and pathological ap- pearances have been found in the lungs of to page 5 The Pathogenesis and Radiological Features of Hut Lung women who have not ground maize and from others living elsewhere in SouthAf- rica.' It should be noted however that unlike exposure in industry, domestic exposure begins at a very young age. ltis possible that previously considered safe doses of silica delivered to infants or adolescent children might be pathogenic. Infants of mothers who grind are frequently bound on the mother's back or lie at her side,and girls take part in the family chores from their early teens. Smoke from bio-mass fuelled fires (coal, wood, dung or plant material) contains car- bon black and a range of other particulate material which is retained in the lungs and is visible bronchoscopically both free within the airways and submucosa. Pathol- ogy specimens confirmed the blackening of lung tissue with focal aggregates of carbonaceous material, and even anthracotic nodules. A common practice in rural traditional dwellings is to cook in- side a hut (rondavel) which has no central chimney, and in cold weather farnilymem- bers sleep around the fire for warmth. A field study has confirmed exceedingly high smoke concentrations under these condi- tions. The role of tuberculosis in patients with . hut lung is difficult to define. A large pro- portion of patients show features compat - ible with healed tuberculosis raising the question of which came first, and whether calcified hilar glands represent silica expo- sure or are the result of tuberculosis infec- tion. It is also possible that, as in gold min- ers pneumoconiosis, patients with hut lung have increased susceptibility to tuberculo- sis. No satisfactory epidemiologic data is available to provide answers to these ques- tions. Pipe-smoking by rural Transkei women is not uncommon and tobacco smoke particles may also playa role in the pathogenesis/ It is therefore likely that hut lung is the result of a mixture of dusts, delivered to the lung in women in pursu- ing a traditional rural lifestyle. The prevalence of hut lung has not been established, but as lifestyle changes occur (including the design of huts, the pro- vision of chimneys in dwellings, changing patterns of cooking and use of fuels, and increased reliance on commercially ground maize), the incidence of this disease can be expected to diminish or disappear. If em- ployed alone, the previously suggested pro- vision of maize grinders at local stores to replace grinding between rocks in the home is unlikely to eliminate this condition. I Clinical presentation Patients are frequently asymptomatic or are X-rayed because of upper respi- ratory tract symptoms relating to acute infections. Many are picked up during contact tradng for tuberculosis. A mi- nority have symptoms compatible with chronic bronchitis with variable degrees of dyspnoea. Severe caseshave features of cor pulmonale. One-third of patients in our series were smokers and a similar proportion had histori- cal or X -ray evidence of previous tu- berculosis, several without microbio- logical confirmation. 2 Chest radiology Chest radiographic changes range from dif- fuse fine rounded regular nodules resem- bling miliary tuberculosis to extensive fi- brosis resernblingprogressive massive fibro- sis (Figures 1a and 2a). The majority have nodules of varying size and density. The profusion scores (ILO-UICC scoring sys- tem for pneumoconiosis) range from 1/1- 3/3 changes. All zones of the lung are usu- ally involved although not uniformly, but progressive massive fibrosis usually affects upper lobes and is surrounded by charac- teristic traction bullae. 5 SA JOURNAL OF RADIOLOGY. May 1996 Figure la: Chest radiograph of Il 59-year old woman with hut lung showing the diffuse fine nodular interstitial patlem with loss of vascular clarity. Figure tb: HRCT lung views of the patient shown in la confirming the presence of evenly distributed centri/obular nodules in profusion. Fissures are nol thickened and bronchovascular markings are normal. H~ resol~tion scanning We have recently investigated the appear- ance of hut lung on high resolution CT scan and have confirmed a high incidence of areas oflocalised scarring compatible with tuberculosis, particularly in the lung apices and frequently associated with calcified hilar adenopathy Egg-shell calci- fication as found in silicosis has not been observed. Localised areas of bronchiectasis suggest previous tuberculous or non-tuber- culous pneumonia complicated by to page 6 The Pathogenesis and Radiological Features of Hut Lung scarring. The varying density and size of the nodules is uniform and indistinguish- able from the appearances of miliary tuber- culosisor of nodular sarcoidosis(Figures 1b and 2b). However, the beaded pattern along bronchovascular markings which is characteristic of sarcoidosis has not been observed. We have concluded however that the CT appearances of hut lung are indistinguishable from these other two conditions,and for this reason HRCT isnot viewed as an essential investigation in such cases. Figure 28: Chest radiograph of a B3-year old woman with hut lung. Nodules are of variable size and less profuse In the apices. Vascular clarity is lost. Figure 2b: HRCT views of the lungs of the patient in 2a. A diffuse eentrilobular distribution is seen with slight bronchial wall thickening. Irregular spiculated nodules of varyIng size are evident, some possibly representing bronchovascular markings. Interlobular fines are thIckened. Diagnosis Fibreoptic bronchoscopy with transbronchial biopsies is of greater value for confirming the diagnosis. Segmental large volume (200ml) bronchoalveolar lavage (BAL) is of some value in dis- tinguishing hut lung from tuberculo- sis and sarcoidosis. In sarcoidosis lymphocyte percentages in the BAL fluid are generally elevated (>Il %). Similar el- evations in lymphocyte numbers are present in a large proportion of patients with miliary tuberculosis, but also in a minority of patients with hut lung. A common feature in hut lung is the large proportion of macro phages that contain in- organic inclusions and appear black under light microscopy. Total cellnumbers are not elevated and the proportions of other inflammatory cells are not increased.' Conclusions The concept of domestically acquired pneumoconiosis is not unique to South Africa,but severalpractices of rural lifestyle commonly practiced in South Africa con- tribute to the epidemic here. These include cooking with bio-mass fuels on open fires in poorly ventilated huts, grinding maize between quartz-containing rocks, pipe smoking and possibly tuberculosis. The incidence of the disease should decline over coming decades, but it remains an important diagnosis to consider in patients with interstitial lung disease,and requires chlferentia- tion from sarcoidosis, miliary tuber- culosis and other diseases. Although most cases are mild, it might predis- pose to tuberculosis, and is often as- sociated with areas ofbronchiectasis which giverise to chronic or intermit- tent symptoms. Respiratory failure and death are probably rare but rec- ognised complications. Although the HRCT has been used to define the 6 SAJOURNAL OF RADIOLOGY· May 1996 spectrum of changes in patients with hut lung, it does not provide an alternative to fibreoptic bronchoscopy for confirming the diagnosis. References 1. Palmer PES, Daynes WG.Transkei silicosis.SAfrMed ]]967;41:1182-8. 2. Grobbelaar JP, Bateman ED. Hut lung: a domesti- cally acquired pneumoconiosis of mixed aetiology in rural women. Thorax 1991;46:334-340. 3. Hirsch M, Bar-Ziv J, Lehmann E, Goldberg GM. Simple siliceous pneumoconiosis of Bedouin females in the Negev desert. Clin Radio1l974;25:507-1 O. 4. Tapp E, Curry A,Anneld c. Sand pneumoconiosis in an Egyptian mummy. BMII975;ii:276. 5. NorbooT,Angchuk PT, Yahya M et al. Silicosis in a Himalayan village population: role of environmental dust. Thorax 1991;46:341-3. 6. Dubovsky H. Pneumoconiosis and tractor plough- ing. Medical Proceedings 1968; 14:56-59. 7. Green FHY, Yoshida K, Fick G, PaulJ, Hugh A, Green WF. Characterization of airborne mineral dusts asso- ciated with farming in rural Alberta, Canada. IntArch Occup Environ Health 1990;62:423-30.