A randomized controlled trial comparing non-steroidal anti-inflammatory and fusion protein inhibitors singly and in combination on the histopathology of bovine respiratory syncytial virus infection

Bovine respiratory syncytial virus (RSV) has substantial morbidity in young calves, and closely parallels human RSV in infants. We performed a randomized controlled trial in five to six-week-old Holstein calves (Bos taurus). comparing fusion protein inhibitor (FPI) and non-steroidal anti-inflammatory drug (NSAID) singly and in combination at three and five days after experimental BRSV infection. Thirty-six calves received one of six treatments; Ibuprofen started on day 3, Ibuprofen started on day 5, FPI started on day 5, FPI and Ibuprofen started on day 3, FPI and Ibuprofen started on day 5, or placebo. We have previously reported significant clinical benefits when combined FPI and NSAID treatment was started at three and five days after bovine RSV infection. Necropsy was performed on Day 10 following infection and hematoxylin and eosin staining was performed on sections from each lobe. Histology was described using a four-point scale. We performed canonical discrimination analysis (CDA) to determine the structural level where differences between treatments occurred and mixed effects regression to estimate effect sizes. Separation from placebo was maximal for dual therapy at the levels of the alveolus, septum, and bronchus in CDA. We found that the clinical benefits of combined FPI and NSAID treatment of BRSV extend at least partially from histopathological changes in the lung when treatment was started three days after infection. We found decreased lung injury when ibuprofen was started as monotherapy on day 3, but not day 5 following infection. Combined therapy with both an FPI and ibuprofen was always better than ibuprofen alone. We did not prove that the clinical benefits seen starting FPI and ibuprofen five days after infection can be solely explained by histopathological differences as identified on H&E staining.


Bronchial mononuclear infiltrates
Bronchial mononuclear infiltrates: Characterized by the presence of lymphocytes, plasma cells and/or macrophages in the bronchial wall. 0: No mononuclear infiltrates. 1: Small numbers of mononuclear cells, above the muscularis mucosa. 2: Small numbers of mononuclear cells above and below the muscularis mucosa. 3: Large numbers of mononuclear cells above and below the muscularis mucosa.

Bronchioles
Score each slide as 0 to 3 for each finding

Bronchiolitis obliterans
Bronchiolitis obliterans: Presence of polyp-like projections of the mucosa that partially or totally occlude the bronchiolar lumen.
0: No bronchiolar polyp-like structure in the section. 1: One bronchiolar polyp-like structure in the section. 2: Two bronchiolar polyp-like structure in the section. 3: Three or more bronchiolar polyp-like structures in the section.

Necrosis of bronchiolar epithelium
Necrosis of bronchiolar epithelium: Characterized by the presence of cells with hypereosinophilic cytoplasm and pyknotic/karyorrhectic nucleus in bronchioles.

Alveolus
Score each slide as 0 to 3 for each finding

Necrosis
Necrosis: Characterized by the presence of cells with hypereosinophilic cytoplasm and pyknotic/karyorrhectic nucleus in alveolar spaces.

Hemorrhages
Hemorrhages: Characterized by the presence of extravasated erythrocytes.

Granulomas
Granulomas: Characterized by the presence of a core of necrosis containing necrotic cellular debris and neutrophils, all surrounded with macrophages, lymphocytes and plasma cells and a peripheral capsule of fibrous connective tissue. 0: No granulomas. 1: One granuloma in the section. 2: Two granulomas in the section. 3: Three or more granulomas in the section.

Vasculitis
Vasculitis: Presence of inflammatory cells in the vascular wall.

Thrombosis
Thrombosis: Presence of a fibrin plug partially or totally occluding the lumen of an alveolar capillary.

Lymphoid nodules
Lymphoid nodules: Presence of nodular lymphoid aggregates in the interstitium.

Septum
Score each slide as 0 to 3 for each finding

Fibrosis
Fibrosis: characterized by the presence of collagen in the pleura.
0: same amounts of collagen (compared to the control) 1: Mild fibrosis (double the amounts of collagen compared to control) 2: Moderate fibrosis (Triple the amounts of collagen compared to control) 4: Severe fibrosis (More than 4 times the amount of collagen compared to control).

Lymphatic dilation/edema
Lymphatic dilation/edema: Lymphatic dilation is characterized by the presence of ectatic lymph vessels. Edema is characterized by the presence of an extracellular eosinophilic proteinaceous amorphous material.  Within the bronchiolar lumen, there is a small deposit of fibrin, which is lined with a few epithelial cells and macrophages. This lesion was interpreted as early bronchiolitis obliterans. Hematoxylin and eosin, 40X.

Data Entry
A sample data entry form printed in spreadsheet software is shown overleaf. This required extensive coding to normalize it and to make it usable for Stata. Ideally, a database rather than excel should be used for data entry tools. However, this tool has the advantage of being extremely e cient for the microscopist and is instantly readable for humans.
For subsequent code the first letter identifies where the sample was taken from, the second part the structure, the third part the specific finding e.g. ABronchussdeciliation refers to slide deciliation graded 0 to 3 in the bronchus from slide A.
See the manuscript and this early working paper see our how gross pathology was incorporated to deal with the discontinuous nature of lung pathology and validated with respect to clinical scores in life.

ssc install plotmatrix
Set the sort order to allow subsequent labelling of individual calves.

CHAPTER 4. PLOTTING THE RESULTS USING MATRIX PLOTS
Additional editing was performed in the Stata graph editor and recorded in the .grec files. These .grec can be edited manually but are verbose and typically re-run as saved via the editor. This summarizes the code. This code was run under Stata 16.1 but will likely work under most recent earlier versions. It will run in future versions by inserting the command.
version 16.1 The following code is unique to this study. It creates and applies labels for subsequent output and converts the some microscopists or data entry clerks habit of leaving normal as blank or placing a check mark instead of writing 0. Here we assume the use of consistent slide nomenclature throughout the analysis as we have done with a diagram of the lungs showing from where each slide is to be taken.