Comparison of bacterial cultivation, PCR, in situ hybridization and immunohistochemistry as tools for diagnosis of Haemophilus somnus pneumonia in cattle

Abstract

The aim of the present study was to compare the potential of bacterial cultivation (BC), PCR, in situ hybridisation (ISH), and immunohistochemistry (IHC) in the diagnosis of Haemophilus somnus, when applied to pneumonic bovine tissue. Lungs from 65 field cases submitted for bacteriological examination were included in the study. The PCR-detection was performed on three different samples: plate-PCR (detection on plate washes after incubation of lung tissue on agar plates); swab-PCR (direct detection on a swab from the cut surface); and, whenever possible, a bronchus-PCR (direct detection on a swab from the main bronchus of the right cranial lung lobe). In order to examine the pathological significance of the findings, a histopathological examination of the cases was performed. H. somnus was detected by one or more techniques in 33 cases in total. By BC the bacterium was isolated from 10 cases, IHC and ISH were positive in 17 and 19 cases, and plate- and swab-PCR were positive in 21 and 29 cases, respectively. The bronchus-PCR was positive in 30 out of 61 cases examined. The PCR-technique was the most sensitive method, and as this technique is fast and relatively inexpensive, it should be considered as a supplementary tool in the diagnosis of H. somnus induced calf pneumonia.

Introduction

Haemophilus somnus is a Gram-negative bacterium capable of inducing a variety of bovine diseases including pneumonia and thrombotic meningoencephalitis (Humphrey and Stephens, 1983). In Denmark, H. somnus is one of the most common bacteria isolated from severe cases of calf pneumonia (Tegtmeier et al., 1999a).

Diagnosis of field cases of bacterial calf pneumonias is normally based on bacterial cultivation. However, bacterial cultivation is suspected to have a low sensitivity, as attempts to isolate pathogenic bacteria from cases of pneumonia often fail, even though their presence might be expected based on clinical and histopathological observations (Tegtmeier et al., 1999a).

In the diagnosis of calf pneumonia, especially isolation of H. somnus can pose a problem. H. somnus is particularly susceptible to antibiotics (Martin and Meek, 1981), and it is recommended that specimens are protected from drying and cultured as soon as possible (Quinn et al., 1994). No or very few live bacteria may otherwise be present, at the time of cultivation, as the bacteria have succumbed during transportation to the laboratory under sub-optimal conditions. Most isolates of H. somnus also demand CO₂ for growth, and grow as tiny pin-point colonies on blood agar plates with variable haemolysis. These colonies can consequently be overgrown due to the post-mortem contaminating bacteria, e.g. Proteus species. Furthermore, isolation of bacteria is often hampered if intensive antibiotic treatment has been given to the animals prior to death. In a recent study, antibacterial substances were detected in 32% of examined bovine lungs (Tegtmeier et al., 1999a), and the difficulty of isolating H. somnus in animals treated with antibiotics has previously been shown (Martin and Meek, 1981). For these reasons, supplementary tools should be considered in addition to cultivation, in order to enhance the detection rate of H. somnus. Today, a variety of methods are available for detection of micro-organisms, including immunohistochemistry (IHC) (Larson, 1989), in situ hybridisation (ISH) (Brown, 1998) and polymerase chain reaction (PCR) (Tang and Persing, 1999). A major advantage in common for ISH and IHC is that a topologic assessment of the examined tissue is possible. This is not the case when detection is made by PCR. The PCR-technique has proven to be a very sensitive diagnostic tool (Angen et al., 1998), but it may be difficult to draw conclusions about the pathological significance of a positive PCR test. Previous studies have evaluated the potentials of such newer techniques in the diagnosis of different micro-organisms in various animals and tissues. These methods have often been found more sensitive as compared to bacterial cultivation (Thoresen et al., 1994; Boye et al., 2000) or virological examination (Burgesser et al., 1999; Larsen et al., 1999). The aim of the present study was to compare the sensitivity of bacterial cultivation, PCR, ISH and IHC in the diagnosis of H. somnus when applied to pneumonic bovine tissue.