OUP user menu

Common variable immune deficiency: respiratory manifestations, pulmonary function and high‐resolution CT scan findings

K.M. THICKETT , D.S. KUMARARATNE , A.K. BANERJEE , R. DUDLEY , D.E. STABLEFORTH
DOI: http://dx.doi.org/10.1093/qjmed/95.10.655 655-662 First published online: 1 October 2002

Abstract

Background: Common variable immune deficiency (CVID) is prone to under‐diagnosis and may not reach relevant specialists until late in life. Morbidity is most commonly due to acute‐on‐chronic respiratory infections leading to respiratory failure.

Aim: To investigate respiratory complications, lung function and high‐resolution computerized tomography scan (HRCT) findings and mortality in 47 patients with CVID.

Setting: A regional immunology unit (Birmingham Heartlands Hospital).

Design: Retrospective observational case‐note study following the introduction of shared care between immunology and respiratory medicine.

Results: Age at diagnosis ranged from 5 to 72 years, with a median time from development of first symptoms to diagnosis of 4.0 years. There was delay in referral between chest physicians and immunologists, (median referral time between specialities >5 years). Forty‐two patients had respiratory complications, due to bronchiectasis (n=32), asthma (n=7), recurrent chest infections (n=9) without concomitant evidence of structural lung damage, and granulomatous lung disease (n=2). Spirometry was abnormal in 10/39 patients (7 obstructive, 3 restrictive). Bronchiectasis was confirmed on chest radiograph (n=9) and HRCT (n=24). Despite the high prevalence of bronchiectasis, few patients had received instruction in physiotherapy and sputum culture results were sparse.

Discussion: To reduce the morbidity associated with CVID, there needs to be greater awareness of respiratory complications, particularly amongst physicians caring for such patients. Emphasis has been placed on adequate dosage of immunoglobulin, but early involvement by a respiratory physician is essential to monitor lung function and initiate optimal therapy, to minimize the occurrence and progression of lung damage.

Introduction

Common variable immune deficiency (CVID) is a heterogeneous syndrome characterized by hypogammaglobulinaemia, recurrent bacterial infections and various immunological abnormalities.1 In addition to recurrent rhinosinusitis, respiratory tract infections, bronchiectasis, liver disease and malabsorption, there is an increased incidence of autoimmune disease, lymphoid proliferation and malignancy. Patients with CVID may therefore present to a variety of medical specialists. There is an average diagnostic delay of 2.5 years in children and 5.5 years in adults.2 This may be due to the misconception that primary antibody deficiency is mainly a disease of childhood, though 95% of patients with CVID present after the age of 6 years, with a mean age of 29 years (range 3–71 years).3 The prevalence of CVID ranges from 1:50 000 to 1:200 000 in Northern European populations.4,,5 However, the UK register for primary immune deficiencies6 contains <1000 cases, whereas there should be around 2500 cases, calculated on the basis of known incidence in similar studied populations elsewhere.7 As few as 1:4 hospital consultants probably have a patient with CVID on their list,7 and it is therefore likely that the condition will often be managed by a clinician with very little prior experience.

Under‐diagnosis, diagnostic delay and management by a variety of medical specialties contribute to morbidity and early mortality. Morbidity is most commonly due to acute‐on‐chronic respiratory infections, leading to respiratory failure,1,3,8,,9 with a mean age at death of 55.4 years in females and 28.8 years in males.3

There are a limited number of studies that have looked at various combinations of respiratory complications, lung function and high‐resolution computerized tomography (HRCT) findings in patients with CVID, and some of these studies are summarized in Tables 1 and 2.

To address the variation in the standards of management of primary antibody deficiencies, the consensus panel for the diagnosis and management of primary antibody deficiencies7 recommended that all patients have a full assessment and follow‐up by a clinical immunologist, and that patients with persistent sputum production should be assessed and managed jointly with a chest physician, to monitor functional impairment and to help prevent progressive lung damage.

During 1997, we developed a shared‐care approach between respiratory medicine and clinical immunology for patients with CVID. We describe respiratory complications, lung function and HRCT findings in 47 patients with CVID out of 66 patients with primary hypogammaglobulinaemia attending a regional immunology unit. Mortality is reviewed retrospectively over a 12‐year period, but rates cannot be calculated, as there is no denominator.

View this table:
Table 1

Summary of studies of patients with common variable immunodeficiency and/or gammaglobulin deficiency

AuthorsPatient numbers(n)Chest diagnoses(%)Numbers with lung function Respiratory/all causes mortality (n)
Cunningham‐CVID103Recurrent bronchitis1009/23
Rundles,3 1989Bronchiectasis21
Cunningham‐Rundles and Bodian,9 1999CVID248Chronic lung disease (with or without bronchiectasis)6813/57 (includes 3 post heart‐lung/lung transplant)
Heart/lung or lung transplantation3
Previous pneumonia193
Hermans et al.,10 CVID50Pneumonia862/12
1976Bronchiectasis28
Hermaszewski CVID240LRTI (with CVID)9723/74
et al.,8 1992XLA44Bronchiectasis18
Thymoma‐associated hypogammaglobulinaemia7Pulmonary fibrosis1
Sole et al.,11 1992CVID15Recurrent pneumonia87
Bronchiectasis53
Bjorkander et al.,12 CVID24Chronic bronchitis77Obstructive8
1984XLA2Bronchiectasis42Restrictive2
Combined2
Watts et al.,13 1986CVID32Bronchiectasis44Restrictive10
Recurrent pneumonia88Obstructive6
Empyaema6Normal9
Lung abscess6
Asthma9
Rosenberg et al.,14 1979CVID17Bronchitis94Moderate‐severe airflow obstruction 16
Previous pneumonia88Within 20±3 years of symptoms
Popa,15 1994CVID or Ig subclass deficiency42Chronic bronchitis79Obstructive26
Bronchiectasis24
Dukes et al.,16 1978CVID55Bronchiectasis38
Thymoma7
Lymphocytic interstitial pneumonitis5
This studyCVID47Bronchiectasis68Mild airflow obstruction58 deaths in 12 years
Asthma15Moderate airflow obstruction2Median age 62.5 years
Recurrent chest infections19Mild airflow restriction23 deaths were due to a respiratory cause
Granulomatous lung disease4Moderate airflow restriction1
Emphysema4
View this table:
Table 2

CT findings in CVID and X‐linked agammaglobulinaemia (XLA)

AuthorsPatients (n)CT findings (%)
Obregon et al.,17 1994Adults, various PIDs (46)Bronchiectasis 35
Diffuse bronchial wall thickening 39
Parenchymal nodules <10 mm 15
Parenchymal lines 9
Mediastinal nodes 7
Manson et al.,18 1997Children (37)Bronchiectasis 24
Feydy et al.,19 1996CVID and XLA (19)Bronchiectasis 42
Bronchial wall thickening 42
Lobar/segmental collapse 37
Interstitial lesions 32
Air trapping 11
Scarring 42
Curtin et al.,20 1991CVID (28), XLA (10)Bronchiectasis 58
Bronchial wall thickening 18
Normal 24
Curtin et al.,21 1995CVID (27), XLA (10)Bronchiectasis 55
Mediastinal lymphadenopathy 41 (in CVID only)
Kainulainen et al.,22 1999CVID(18), XLA (4)Bronchiectasis 73
This studyCVID (47) (CT in 34)Bronchiectasis 51
Lobar collapse 6
Mediastinal/hilar lymphadenopathy 2
Emphysema 4
Fibrotic change 2
Pleural thickening 6
Cavitating lesion 2
Bronchial cyst 2
  • PIDs, primary immunodefiency disorders.

Methods

During 1997/1998, patients with CVID attending the regional immunology clinic were referred to a respiratory physician with an interest in chronic suppurative lung disease. Information was collected retrospectively from case records regarding exacerbations, frequency of antibiotic usage, sputum microbiology, pulmonary function tests and thoracic HRCT scans. Duration of symptoms prior to diagnosis, previous contact with a respiratory physician, and time taken to first referral to a respiratory physician were noted.

Results

Patient presentation and referral times

Age at presentation, median time to diagnosis from development of first symptoms, and median referral time between respiratory medicine and immunology are shown in Table 3. Patient characteristics, including presenting complaint, clinical diagnosis relating to the respiratory system and smoking status, are shown in Table 4. Only 5/47 patients were free of respiratory complications. Four patients had never been referred to a respiratory physician, of whom one had respiratory symptoms. Only six patients had received instruction in chest physiotherapy. Median body mass index (BMI) was 22.3, (range 16.8–33.0) indicating that for most, nutritional status was not a problem.

View this table:
Table 3

Patient characteristics and referral times between immunology and chest medicine

Patient characteristicsMaleFemaleAll patients
Number (n)272047
Median age [range] (years)39.0 [22–59]55.5 [39–81]45.5 [22–81]
Median age at diagnosis [range] (years)30.0 [5–56]47.0 [13–72]35.0 [5–72]
Median time from first symptoms to diagnosis [range] (years)3.0 [0.8–30]7.0 [1–45]4.0 [0.8–45]
Median time for referral from immunology to chest physician [range] (years)8 [0–28]2.5 [0–35]6.0 [0–35]
Median time for referral from chest physician to immunology [range] (years)8 [8]2 [1–12]5.0 [0–12]
View this table:
Table 4

Presentation, respiratory diagnosis, respiratory infection, smoking status and antibiotic usage

Characteristicn (%)
Presenting complaint
Recurrent chest infection26 (55)
Productive cough19 (40)
Wheeze4 (9)
Weight loss4 (9)
Rhinosinusitis3 (6)
Otitis media3 (6)
Respiratory clinical diagnosis
Bronchiectasis32 (68)
Recurrent chest infection/recurrent pneumonia9 (19)
Asthma7 (15)
Granulomatous lung disease2 (4)
Emphysema*2 (4)
Previous tuberculosis1 (2)
Cavitating lung lesion1 (2)
Rhinosinusitis22 (47)
Sputum microscopy
Haemophilus influenzae**15 (32)
Streptococcus pneumoniae 4 (9)
Haemophilus parainfluenzae 1 (2)
Moraxella catarrhalis 1 (2)
Methicillin‐resistant Staphylococcus aureus1 (2)
Normal flora3 (6)
Alcohol acid‐fast bacilli (AAFB)***0 (0)
No sputum result available22 (47)
Antibiotic usage
Continuously7 (15)
Monthly rotating antibiotics13 (28)
As required27 (57)
Smoking status
Current smoker8 (17)
Ex‐smoker6 (13)
Never smoked30 (64)
Unknown3 (6)
  • *One patient age 33 had extensive emphysema with a normal α‐1 antitrypsin level and had never smoked. **14 patients who had positive sputum culture for Haemophilus influenzae and 2 patients with positive sputum culture for Streptococcus pneumoniae had HRCT evidence of bronchiectasis. ***9 of 47 patients had sputum test for AAFB.

Intravenous immunoglobulin replacement

The majority of patients (24) were receiving 400 mg/kg/month intravenous immunoglobulin (IVIG), 16 received 200 mg/kg, three received 600 mg/kg and one each received 300 mg/kg and 500 mg/kg, respectively.

Respiratory infections

Antibiotic usage and sputum microscopy results are shown in Table 4. No sputum culture results were available in 22 patients. The majority of patients who had positive sputum culture results had evidence of bronchiectasis on HRCT. Respiratory tract infection (manifest by increase frequency of cough, change in sputum purulence or requirement for antibiotics) was documented in the notes of 34 of the 47 patients, ranging from 0–12 infections per year with a median of 2.5 infections per year. Nine patients had negative sputum results for alcohol acid‐fast bacilli (AAFB); in the remaining 38, no result was recorded. There was no evidence that any had had tuberculosis.

Pulmonary function

Spirometry was normal in the majority of patients (29 of the 39 who had spirometry performed)—seven had airflow obstruction (FEV1 <80% predicted) and three had a restrictive pattern (Table 5). Maximum mid‐expiratory flow rate (MMEF), which measures the average slope of the expiratory curve between 25–75% of vital capacity, was abnormal in 13 of the 19 patients in whom it was measured.

View this table:
Table 5

A comparison of chest radiograph and HRCT findings

AbnormalityChest X‐ray (n=33/47) (n)HRCT(n=34/47) (n)
Bronchiectasis924
Lobar collapse13
Mediastinal/hilar lymphadenopathy01
Emphysema22
Pleural thickening33
Cavitating lesion11
Bronchial cyst01
Lobectomy01
Normal148
Peribronchial shadow10
Bilateral lower zone shadows20
Fibrotic change01
  • Not all patients had a HRCT scan and a chest radiograph.

Radiology

The chest radiograph and HRCT results are shown in Table 6. One patient appeared to have a marked reduction of peribronchiectatic shadowing on HRCT taken a year later after instruction in physiotherapy and intensive antibiotics. Two patients were thought to have granulomatous lung disease, one of whom was confirmed by transbronchial biopsy; the other had a HRCT with hilar and mediastinal lymphadenopathy.

The HRCT findings in patients with an abnormal MMEF yet otherwise normal spiromerty were reviewed. Four had a normal HRCT, two had evidence of bronchiectasis following HRCT and one patient did not have a HRCT scan, although a chest radiograph showed bronchiectasis. All seven patients had respiratory symptoms.

View this table:
Table 6

Lung function in 39 patients with gammaglobulin deficiency

FEV1FVCMMEF 25–75TLCRVTLCOKco
All39391919191813
>80% predicted2934617151311
60–79% predicted7*552330
40–59% predicted3**030111
<40% predicted0050011
  • Data are numbers. FEV1, forced expiratory volume in 1 s; FVC, forced vital capacity; MMEF, maximum mid expiratory flow rate; TLC, total lung capacity; RV, residual volume; TLCO, transfer factor; Kco, transfer coefficient. *5 obstructive, 2 restrictive; **2 obstructive, 1 restrictive.

Serum angiotensin converting enzyme

Serum angiotensin converting enzyme (SACE) was measured in eight patients (range 26–113 U/l) (normal range 8–59 U/l). Two patients had raised SACE levels. One patient with a SACE of 103 U/l had previously had sarcoid confirmed on transbronchial biopsy, and one patient with a SACE of 113 U/l had a normal HRCT thoracic scan.

Mortality

Mortality was reviewed over a 12‐year period between 1988 and 2000 (Table 7). The number of patients with CVID who had attended the immunology department over that period was not definitely known, but a search of all known records over the period revealed that eight patients (5 male, 3 female) had died, with an median age at death of 62.5 years (range 38–76 years). Median age of death in females was 60 years (range 53–76 years); that in males was 47 years (range 38–73 years). Three patients (37.5%) died with a respiratory condition as the primary cause of death (median age 47, range 42–76 years), and in a further two (25%), respiratory disease was a contributory cause. One patient had received a heart lung transplant for respiratory failure thought to be due to a combination of bronchiectasis and emphysema. The explanted lung histology showed granulomatous disease and obliterative bronchiolitis. He died from respiratory failure 30 months post transplant, and the post‐mortem examination confirmed granulomatous disease in the transplanted lung.23

View this table:
Table 7

Retrospective causes of mortality in eight patients attending the immunology clinic reviewed over 12 years

Age (years)Sex (M/F)Primary cause of deathComorbidities
53FCarcinoma of the colonBronchiectasis
60FCerbrovascular accident
76FPneumonia
Bronchiectasis
38MCarcinoma of the rectumCoeliac disease
42MMyocardial infarction
Respiratory failure
Emphysema
47MObliterative bronchiolitis 30 months post heart‐lung transplant. Post‐mortem histology showed granulomatous lung diseaseIschaemic heart disease
65MCarcinoma of the oesophagusBronchiectasis
73MMyocardial infarction

Discussion

The median age at diagnosis for males in this series was similar to that of Cunningham‐Rundles and Bodian in the US.9 Our median age at diagnosis for females was higher (46.9 vs. 33 years). This might relate to a difference in disease severity between males and females (male patients with CVID appear to die earlier than females3,,9) and increased median delay in diagnosis in our females (7.0 years for females vs. 3.0 years for males), whereas in the series by Cunningham‐Rundles and Bodian,9 there was a delay of 6 years for males and 5 years for females. Diagnosis of CVID is often delayed, the median time to diagnosis from developing first symptoms for all patients in this study being 4.0 years. There was often a long delay in referral, with a median referral time overall of over 5 years, between both respiratory physicians and immunologists in each direction.

There is considerable morbidity and mortality. In this study, the most common presentation was recurrent chest infection and productive cough. 42 out of 47 patients had respiratory symptoms mostly due to bronchiectasis, asthma and recurrent chest infections. One patient had advanced emphysema on HRCT, despite being a non‐smoker and having a normal α‐1 antitrypsin level. Rosenberg et al.14 reported three cases of emphysema with a predilection for the lung bases similar to the appearance of homozygous α‐1 antitrypsin deficiency, although this was not found.

Death is most commonly due to acute or chronic respiratory disease or malignancy.1,3,8,,9 In the series of 103 patients followed for up to 13 years by Cunningham‐Rundles3 (Table 1), 23 died at a young age, with a mean age at death of 28.8 years in men and 55.4 years in women. Nine died from respiratory failure. In a more recent study of 248 patients with CVID (Table 1) Cunningham‐Rundles and Bodian9 found that 57 patients had died after a median follow‐up period of 7 years, with a mean age at death of 40 years in males and 45.5 years in females. Whilst the single major cause of death was lymphoma (n=10), respiratory disease was another major cause, including chronic pulmonary infections resulting in cor pulmonale (n=6), respiratory failure (n=4), chronic or acute rejection post lung transplantation (n=3), Pneumocystis carinii pneumonia (n=1) and lung cancer (n=1). In our series, a large proportion (37.5%) died of malignancy, but similar to results obtained by Cunningham‐Rundles,3,,9 over a third of our deaths had a respiratory cause, with a median age at death of 62.5 years.

Respiratory physicians need to be vigilant for hypogammaglobulinaemia in any patient with unexplained recurrent respiratory infections or rhinosinusitis, which may be diagnosed by low total immunoglobulins or the demonstration of poor functional antibody responses to antigens encountered through natural exposure or immunization (for example, antibodies to tetanus, diphtheria, measles, mumps, rubella and capsular polysaccharide antigens of Haemophilus influenzae type b and Streptococcus pneumoniae).

Despite the high prevalence of bronchiectasis, very few of our patients had ever received instruction in airway clearance techniques. The majority of patients had had two or more courses of oral antibiotics per year for chest infections, although in our view antibiotics were not always the most appropriate, and dosages were suboptimal for this group of patients. Sputum microscopy and culture results were seldom recorded, even though they may be useful in detecting infections which may require antibiotics. The duration of antibiotic treatment was not clear in all cases, but a prescription in favour of conventional short courses may be inappropriate for bronchiectasis. Many patients self‐administer their IVIG at home and therefore would also be familiar with the techniques necessary for self‐administration of intravenous antibiotics.

In studies by Watts et al.13 and Popa,15 up to half the patients had abnormal spirometry, and 10/39 patients in our study had spirometric abnormalities. Maximal mid‐expiratory flow (MMEF) was abnormal in 13 patients, of whom seven had normal spirometry and three had a normal HRCT. In cystic fibrosis, small‐airways disease develops early in the disease course, and changes in MMEF precede other changes in lung function.24 Similarly, a decrease in MMEF may be an indicator of early airways disease in the smallest airways or bronchioles in CVID, as it is in bronchiectasis from all causes.

Although there has been a recent emphasis on the role of higher‐dose (600 mg/kg rather than 200 mg/kg) IVIG in reducing acute infections and improving lung function and improvements in chest radiology,25 other aspects of management also need to be considered. Recently, Kainulainen22 demonstrated silent progression of bronchiectasis in 5/14 patients on HRCT scan over 3 years despite the maintenance of preinfusion IgG concentrations of 5 g/l or more.

During the last 50 years, the median survival of patients suffering from cystic fibrosis has increased from 5 to 32 years.26 This has been largely due to improvements in antibiotics, airway clearance techniques and nutrition. CVID is generally similar to cystic fibrosis in its consequences for the exacerbation of pulmonary pathology, and the sequelae are similar, namely chronic pulmonary sepsis, bronchiectasis, chronic rhinorhinosinusitis and sometimes malabsorption. Recognizing this similarity to patients with cystic fibrosis, the early involvement of a respiratory physician would seem desirable, to monitor lung function and optimize chest care, including optimal antibiotic treatment, good inhaler techniques and smoking cessation. Physiotherapy advice regarding which techniques to adopt and to optimize anatomical lobar drainage and supervision of airway clearance methods before irreversible structural lung damage occurs is desirable.

In addition to optimal immunoglobulin dosing the use, duration and dosage of antibiotic treatment or prophylaxis should be considered for each patient.

The role of lung transplantation in CVID patients with respiratory failure needs to be addressed. The known experience is limited.9,23,,27 Successful lung and liver transplantation has been reported27,,28 but long‐term follow‐up is not available. In the study by Cunningham‐Rundles and Bodain,9 two patients aged 26 and 27 years, respectively, underwent bilateral lung transplant and heart‐lung transplant, both for respiratory failure, and remained well for 2 and 3 years before developing progressive chronic rejection and an undiagnosed bleeding disorder characterized by platelet dysfunction and pulmonary bleeding. A further patient aged 32 years received a bilateral lung transplant for granulomatous lung disease but died 3 days post transplantation following surgical complications. Chronic immunosuppression in CVID for autoimmune conditions has been associated with complications, including Pneumocystis carinii pneumonia, Nocardia brain abscess, progressive multifocal leukencephalopathy, and recurrence of Hodgkin's disease,9 and the role of long‐term immunosuppression post transplantation in CVID is thus currently uncertain.

To improve morbidity and mortality in CVID, there needs to be heightened awareness amongst physicians and involvement of respiratory expertise at an early stage. Regular, and preferably joint follow‐up by both specialities will enhance optimization of IVIG dosage, allow early identification of respiratory tract infections, enable detection of airway disease by flow volume loop measurement, allow observation of recovery by serial spirometry, and provide a logical antibiotic choice and dosing regimen based on microbiological findings. Hence the development of irreversible pulmonary changes may be limited and early mortality due to lung disease prevented.

Footnotes

  • Address correspondence to Dr D.E. Stableforth, Adult Cystic Fibrosis Unit, Department of Respiratory Medicine, Birmingham Heartlands Hospital, Bordesley Green East, Birmingham B9 5SS

References

View Abstract