Introduction

The term pneumothorax was first coined in 1803 by a medical student, Itard, while describing a collection of air within the pleural space. (₁) Most pneumothoraces occur after thoracic trauma (traumatic) or after medical interventions involving the thorax (iatrogenic). A pneumothorax that occurs in an individual without any clinically apparent predisposition is termed a spontaneous pneumothorax.

There are two recognized types of spontaneous pneumothorax. Primary spontaneous pneumothorax (PSP) occurs in individuals without clinically evident underlying lung disease whereas secondary spontaneous pneumothorax (SSP) occurs as a complication of pre-existing lung pathology. (₁,₂,₃).

Spontaneous pneumothoraces are relatively uncommon, with reports of the overall incidence (combined SSP/PSP) ranging from 7.8 (₂) to 16.9 (₃) per 100,000 persons per year. There has been no report on the prevalence of this disease in the Caribbean. We sought to report on the prevalence of spontaneous pneumothorax in the setting of a developing Caribbean nation.

Methods

There are an estimated 826,880 persons residing in and around Kingston, the capital of Jamaica. (₄) When patients within this population develop pneumothoraces, they are usually sent to tertiary referral centres in Kingston for definitive management.

There are three tertiary referral hospitals in Kingston: The Kingston Public Hospital and the University Hospital of the West Indies are tertiary level hospitals that provide multidisciplinary care to unselected patients. (₅) The National Chest Hospital is a specialist hospital that provides sub-specialty care to the select patient population with respiratory illnesses. (₅)

The admission records from these three institutions were reviewed over a five year period between January 1997 and December 2001. All patients who were treated for pneumothoraces during the study period were identified and their hospital records were retrieved for analysis. Patients who had pneumothoraces without a clinically apparent reason were considered to have spontaneous pneumothoraces. We excluded patients with traumatic and iatrogenic pneumothoraces from further analysis.

The following data were retrospectively extracted from the hospital records: patient demographics, presence of co-morbidities, smoking history, presence of associated lung pathology, location and estimated size of the collection, treatment details and therapeutic outcomes. The extracted data were entered into a Microsoft Excel worksheet and analyzed using SPSS version 12.0.

The episodes were classified as primary spontaneous pneumothoraces (PSP) if there was no evidence of underlying lung disease and secondary spontaneous pneumothoraces (SSP) if the condition occurred as a complication of underlying lung disease. Patients who had a confirmed prior admission for SP were considered to have recurrent disease.

Results

There were 81 patients treated for spontaneous pneumothoraces over the study period. The overall incidence for all cases of spontaneous pneumothoraces in this setting was 1.96 per 100,000 population per year. There was a preponderance of males with an overall male to female ratio of 3:1. The majority of patients (45%) were above the age of 50 years (Figure 1).

Figure 1

There were 23 patients (28.4%) with SSP. The underlying lung disorders are listed in Table 1. There were 8 females with an average age of 51.4 +/-21.9 years and 15 males with a mean age of 60.4 +/-15.9 years (mean +/-SD). The incidence of SSP in this setting was 0.36/100,000 per year in males and 0.19/100,000 per year in females (overall incidence of SSP was 0.56/100,000 population per year).

Figure 2

The remaining 58 patients were considered to have PSP. In this group there were 48 males with an average age of 46.7 +/-14.8 years and 10 females with an average age of 40.8 +/-11.6 years (mean +/-SD). The incidence of PSP in this setting was 1.40/100,000 persons per year (1.16 in males and 0.24 in females).

The majority of patients (58/81, 72%) were current smokers of cigarettes and/or cannabis (Figure 2). A history of current smoking was significantly higher among males (83% vs 33%; X ² = 16.67; p <0.001). There were 46 current smokers in the PSP group (39 males, 7 females) and 6 current smokers in the SSP group (5 males, 1 female).

Figure 3

All the patients presented to hospital complaining of sticking chest pain and varying degrees of shortness of breath. One patient, a 35 year old man, was clinically diagnosed with a spontaneous tension pneumothorax within minutes of presentation. The remaining patients had definitive diagnoses made upon plain chest radiography.

There were 13 patients (16%) with recurrent pneumothoraces. Nine were ipsilateral, all on the right and 4 were contra lateral recurrences. The patients were all males between the ages of 45 and 55 years with COPD (9), Tuberculosis (2) and asthma (2). The remaining 68 patients presented to hospital without any previous episodes of spontaneous pneumothorax.

There was a significant predilection for the right hemi-thorax to be affected, with 52 patients (64%) having a right sided pneumothorax. The left side was affected in 28 patients (34%). There was a single reported case of bilateral pneumothoraces in a 92 year old woman with COPD. She had small collections bilaterally, both estimated to occupy less than 15% of the respective hemi-thorax. She was treated definitively with repeated needle thoracostomy during in-hospital observation and surveillance with serial chest radiographs. Her recovery was uneventful.

The sole patient with a tension pneumothorax was treated with a needle thoracostomy at the second right inter-space for emergent decompression. This was followed by definitive treatment with a basal thoracostomy tube.

The remaining 79 patients had basal thoracostomy tubes inserted as the first therapeutic manouevre. The thoracosotomy tubes served as definitive treatment in 74 cases (89%) while 8 patients (9%) had tube thoracostomy followed by pleurodesis.

Pleurodesis was offered to 8 patients, all with recurrent pneumothoraces. Five of these patients had open pleurodesis with gauze abrasion and the remaining three had talc pleurodesis through a thoracostomy tube.

Discussion

The overall reported incidence of spontanteous pneumothorax varies from 7.8 (₂) to 16.9 (₃) per 100,000 persons per year. The preponderance of affected males in this series was in keeping with the published demographics (₃,₆,₇) but there was a lower than expected overall incidence of spontaneous pneumothorax in Jamaica (1.96/100,000 population yearly) compared to international figures.(₁,₃,₆,₇)

This was an unexpected finding, since it has been shown that 36% of men and 11% of women in the general Jamaican population regularly smoke cigarettes and/or marijuana. (₈) Smoking is a recognized risk factor for spontaneous pneumothorax and has been estimated to increase the risk 22 fold in men and 9 fold in women. (₉) The predominant theory is that pollutants from cigarette smoke stimulate the influx of acute inflammatory cells, disrupting the delicate balance between oxidant / antioxidant systems. This eventually leads to the destruction of lung parenchyma and alveolar rupture. (₁₀,₁₁,₁₂)

There were no cases of spontaneous pneumothorax occuring in patients below the age of 18 years. This was not surprising since there are only few sporadic reports of paediatric patients being affected with spontaneous pneumothoraces. (₃,₁₀,₁₃,₁₄,₁₅)

The yearly incidence of PSP is estimated to range from 7.7 (₁₆) to 18 (₁₀) cases per 100,000 population among men and from 1.3 (₁₆) to 6.0 (₁₀) cases per 100,000 population in women. The incidence in this series was lower than expected (1.16/100,000 in males; 0.24/100,00 in females).

PSP typically occurs in tall asthenic males in the second and third decades of life (₅,₆,₁₀) and is strongly related to a history of cigarette smoking, increasing the relative risk of PSP by as much as 20 times in a dose-dependent manner. (₉) In our series, PSP occurred predominantly in males (4.8:1) but the average age of presentation (46.7 +/-14.8 years) was greater than is typically reported. Unfortunately, we were unable to assess the correlation between body habitus and PSP in this population because there was poor documentation of patient height and weight in the hospital records.

As expected, the majority of patients (81%) were current smokers. Although these patients do not have clinically apparent lung disease, it is recognised that over 75% of these patients have ruptured subpleural bullae that have caused PSP. (₉,₁₀,₁₇,₁₈,₁₉) Unfortunately we were unable to evaluate the prevalence of bullous disease in these patients because they were all treated with tube thoracostomy alone and chest computed tomography scans were either not available or not performed in the majority of cases.

SSP is considered to be a more serious disorder because these patients have premorbid cardio-respiratory compromise from the underlying lung disease. The incidence of SSP is similar to PSP, but it typically occurs in the sixth to seventh decades of life because this is the population that is prone to underlying chronic lung disorders. (₃,₁₀) A similar pattern was seen in our series, where SSP occurred at an average age of 51.4 years in females and 60.4 years in men. However, the yearly incidence of SSP was lower than expected in this population (0.36/100,000 in males and 0.19/100,000 in females).

There are several underlying disorders that are recognized causes of SSP, including COPD, Pneumocystis Carinii pneumonia (PCP) in HIV positive patients, malignancies, bronchial asthma, granulomatous lung diseases and cystic fibrosis. (₁₀) Their incidence varies widely depending upon the prevalence of these diseases in the population studied. (₃,₉,₁₀) The underlying lung disorders encountered in this series was comparable to reports from other Western countries with COPD being the most common cause of SSP. (₃,₆,₇,₉,₁₀)

There was a significantly higher incidence of SSP secondary to tuberculosis (26%) in Jamaica. We believe that many of these represent pulmonary complications of HIV infection, although this was not always objectively determined during the patients' admissions for spontaneous pneumothorax. The Pan American Health Organization biostatistics rank the Caribbean second only to Sub-Saharan Africa in terms of the number of HIV cases per capita, with 2.4% of Caribbean adults living with HIV. (₂₀)

The close link between the increasing worldwide incidence of tuberculosis and HIV infection has been well described. (₂₁,₂₂,₂₃) The HIV-induced decline in cell mediated immunity increases the risk of contracting tuberculosis and disease progression in coinfected individuals. (₂₁,₂₂) The increased prevalence of tuberculosis and P. carinii pneumonia in affected individuals can explain the 2-6% prevalence of spontaneous pneumothorax that is reported in HIV-infected patients. (₂₄,₂₅,₂₆)

The clinico-pathologic pattern encountered in this setting is similar to that reported in the medical literature. The prediliction for the right hemi-thorax being affected is consistent with other reports. (₁₀₂₇₂₈)

We encountered a single case of bilateral disease (1.2%), again in keeping with the medical literature where bilateral disease has been reported in 5% of patients with spontaneous pneumothoraces. (₂₇)

A spontaneous pneumothorax is a potentially dangerous event because it results in a decreased vital capacity and increased alvoelar-arterial oxygen gradient, eventually leading to hypoxemia as a result of low ventilation–perfusion ratios and shunting. An early diagnosis is often difficult to ascertain because the clinical features are often subtle as patients can compensate for small early pneumothoraces. Larger progressive collections can severely compromise respiratory reserves and lead to sudden dramatic clinical deterioration. This reinforces the need for early diagnosis and treatment.

The subtle clinical presentation seen in this series was typical of other reports. (₁,₃,₁₀,₂₈,₂₉,₃₀) Most patients with spontaneous pneumothoraces presented without a clear precipitating event (₂₈) and complained of mild ipsilateral chest pain (₁₀,₂₉) and/or dyspnoea. (₁₀,₃₀) These patients compensated for the pneumothorax well, most having the diagnosis cemented on chest radiographs.

There was one episode (2.4%) in our series of a patient presenting with a potentially life threatening tension pneumothorax. This presentation is uncommon, and has been reported to occur in 3% of patients with spontaneous pneumothoraces in the medical literatre. (₂₇)

Conclusion

The clinical features of an early spontaneous pneumothorax are often quite subtle, making it difficult to make the diagnosis. Nevertheless, early diagnosis is paramount to prevent life threatening complications.

The incidence in Jamaica is low and the local demographics differ from reports in the medical literature. Therfore, physicains must remain aware of the regional demographics of this condition and have a high threshold to diagnose spontaneous pneumothoraces in order to deliver effective care.

Correspondence to

Dr. Shamir Cawich Department of Basic Medical Sciences The University of the West Indies, Mona, Kingston 7, Jamaica, WI. E-mail: socawich@hotmail.com