Impact of meloxicam on respiratory virus titers and health outcomes when administered concurrently with a modified live respiratory vaccine in abruptly weaned beef steers

Abstract

Abruptly weaned crossbred steer calves (N = 271) were used in a randomized, blinded 2-arm clinical trial to assess the impact of a long-acting non-steroidal anti-inflammatory drug on bovine herpesvirus type 1, bovine respiratory syncytial virus, parainfluenza virus type 3, and coronavirus titers and health outcomes when administered concurrently with a modified live respiratory vaccine upon arrival at a feedlot. Treatment groups included a control (saline; n = 135) and an experimental group (injectable meloxicam; n = 136). Viral antibody titers and body weight were measured on arrival, day 7, and day 21, along with a final weight on day 45. Body weight and antibody titers for all viruses increased over time (P < 0.001); however, there were no differences by treatment group or a significant group × time interaction when evaluated using repeated measures analysis of variance. Interestingly, the use of meloxicam was associated with increased treatment risk (P < 0.05). In conclusion, the administration of meloxicam may adversely affect health; however, a decreased vaccine response is likely not a contributing factor.

Introduction

Anti-inflammatories are highly recommended as a method to reduce pain and inflammation associated with certain diseases, injuries, stressful events, and painful procedures in beef cattle (1,2). Unfortunately, glucocorticosteroids have been associated with adverse health effects when used as ancillary therapy for diseases such as bovine respiratory disease (BRD) (3). As such, the use of less potent nonsteroidal anti-inflammatory drugs (NSAIDs), such as meloxicam, have gained popularity in recent years (4).

Although it is effective at mitigating pain (2), meloxicam has impacts on health that are not fully understood. When administered to cattle before transport or in conjunction with a lipopolysaccharide, meloxicam failed to mitigate the negative impacts of stress on inflammatory mediators (5,6). In regard to early vaccine response, meloxicam provided no significant improvement when bovine herpesvirus type 1 (BHV-1), bovine respiratory syncytial virus (BRSV), bovine parainfluenza virus type 3 (BPIV-3), and bovine viral diarrhea virus (BVDV) antibody titers were evaluated (7). These small-scale studies taking an in-depth look at cell-mediated and humoral immune responses in healthy well-vaccinated yearling cattle may not have adequate power to detect differences in vaccination response or health outcomes. Furthermore, they may not reflect the complex immunological status of abruptly weaned beef calves. This is of concern in a feedlot setting where cattle may be receiving multiple pharmaceutical products simultaneously and are at high risk for developing BRD.

The aim of this study was to evaluate the impact of meloxicam on vaccination response and health outcomes on a larger scale in a field setting (e.g., a large pen of abruptly weaned beef calves at a commercial feedlot). More specifically, the first objective was to determine the effects of meloxicam administered upon arrival at the feedlot on primary vaccination response. This was achieved by measuring BHV-1, BRSV, BPIV-3, and coronavirus (CV) titers i) during vaccination upon arrival at the feedlot to determine baseline titers, ii) after 7 d to detect acute immune response, and iii) after 21 d to detect peak immune response. The second objective was to determine the impact of meloxicam on health outcomes by evaluating morbidity and mortality through day 45.

This research is intended to benefit the beef industry by enabling producers and veterinarians to make an educated decision about the use of meloxicam in calves at weaning by balancing the need for pain mitigation with potential impacts on immune function and health.

Materials and methods

Study design

This randomized blinded 2-arm clinical trial was conducted in accordance with guidelines established by the Canadian Council on Animal Care. A single large commercial feedlot inducting and processing approximately 250 ranch-direct, abruptly weaned steer calves in 1 d during fall 2018 was recruited for this study so that vaccination history could be collected. There were 2 sources of steers for this study, both located ~2 h from the feedlot. One source supplied 98 head and the other source supplied 174 head of 6- to 9-month-old British (Red Angus) and Continental (Charolais and/or Simmental) crossbred steers to the feedlot. These herds were on similar vaccination and herd health programs, as both groups of steers qualified for the same producer loyalty program and had received a single 5-way modified live vaccine and 7-way clostridial vaccine during April 2018.

At the initial feedlot processing event on October 26, 2018, each steer was uniquely identified by ear tag and information such as body weight (BW), castration and horn status, and other characteristics were recorded using proprietary health management software. A single bull from the larger group requiring castration was excluded from enrollment in the study. A venous blood sample was collected into a serum separator tube and refrigerated pending processing. All 271 enrolled steers received Vista Once (Merck Animal Health, Kirkland, Quebec) subcutaneously (SC) as their 5-way injectable modified live respiratory vaccine, Bovilis Coronavirus (Merck Animal Health) as their coronavirus intranasal vaccine, and Vision 8 Somnus with SPUR (Merck Animal Health) SC as their 8-way clostridial vaccine. They also received a Ralgro implant (Merck Animal Health), Bimectin Pour-On Dewormer for Cattle (Bimeda, Cambridge, Ontario), Safe-Guard Suspension oral dewormer (Merck Animal Health), and Zuprevo (Merck Animal Health) SQ as their macrolide antibiotic.

Steers were blocked by source and randomly allocated to either the control group (saline; n = 135) or the experimental group (injectable meloxicam; n = 136) using a random integer generator (Excel; Microsoft, Redmond, Washington, USA). Rheumocam injectable meloxicam (Merck Animal Health) labeled “for relief of pain following de-budding of horn buds in calves < 3 mo of age” was administered according to the label recommendation of 0.5 mg/kg BW SC (20-day meat withdrawal) to steers in the experimental group. Saline was administered at the equivalent dosing volume. Opaque bottles labeled either “A” or “B” with automatic dosing syringes were used to administer meloxicam and the placebo product to ensure the investigators and staff associated with the study were blinded.

All steers were housed in a single large open feedlot pen with windbreaks and concrete bunks, typical of western Canada. At 7 and 21 d, all steers were re-handled, and subsequent venous blood samples were collected.

Steers were monitored daily throughout the entire study period by feedlot staff. Clinical signs of disease were recorded, and steers were treated according to a protocol developed by the consulting veterinarian. The presence of respiratory signs, depression, and/or anorexia, in addition to increased rectal temperature [≥ 105°F (40.5°C)] was used to diagnose BRD.

Treatment records (product, dose, route, etc.) were logged into a proprietary health management software program by the feedlot staff. All steers that died were necropsied by a veterinarian. Health and performance data were collected at the time of revaccination and re-implant on day 45.

Laboratory analyses

Whole venous blood samples were centrifuged for 20 min and a minimum of 1 mL was frozen and submitted to a regional diagnostic laboratory (Prairie Diagnostic Services, Saskatoon, Saskatchewan) pending analysis. There was no plate-to-plate variation reported, as all 3 sets of samples were analyzed simultaneously. Antibodies to BHV-1, BRSV, BPIV-3, and CV were quantified using an enzyme-linked immunosorbent assay (ELISA) test (8). Well plates were coated with viral antigens purified from both virus-infected and non-infected cell lysates. Serum was diluted to a 1:40 ratio and applied to well plates. Following incubation and washing, bound antibody was detected using horseradish peroxidase conjugated recombinant protein G and o-phenylenediamine dihydrochloride substrate. The result, expressed as ELISA units, represents the ratio of the net optical density of the sample and a positive control multiplied by 100. An increase of ≥ 20 units between time points indicates seroconversion. Due to the potential of maternal interference and expense, BVDV titers were not measured (9).

Statistical analyses

SPSS 24 (IBM, Armonk, New York, USA) was used to analyze all data and significance was established at P < 0.05. As steers from both sources shared a similar arrival BW, breed composition, transportation history, and vaccination history, and were to be housed in the same pen, they were combined for all analyses. Descriptive statistics were performed, and all continuous variables were assessed for normality using the Shapiro-Wilk test. Associations between parametric and nonparametric continuous variables were assessed using a Student’s t-test and Mann-Whitney U-test, respectively. A Chi-square test was used to determine associations between categorical variables. Differences in the direction and magnitude of change of BW and respiratory titers within and between groups were determined using repeated measures analysis of variance (ANOVA). Thresholds for P-value significance were adjusted for multiple comparisons using the Bonferroni method (P < 0.0167; 10). Mauchly’s test was used to evaluate sphericity. When the assumption of sphericity was violated, the P-value from the Greenhouse-Geisser or Huynh-Feldt tests were used when epsilon was < 0.75 or > 0.75, respectively.

Results

All antibody titer data were non-normally distributed. There were no differences by treatment group nor was there a treatment group × time interaction present when BHV-1, BRSV, BPIV-3, and CV antibody titers were evaluated using repeated measures ANOVA (Figure 1). There was, however, a difference in the direction and magnitude of antibody titers for each virus over time (P < 0.001).

Figure 1.

Bovine herpesvirus type 1 (BHV-1), bovine respiratory syncytial virus (BRSV), bovine parainfluenza virus type 3 (BPIV-3), and coronavirus (CV) antibody titers at each handling event by treatment group. Error bars represent a 95% confidence interval. Significance for repeated measures analysis of variance results was P < 0.0167 following the Bonferroni correction.

The proportion of steers that seroconverted (≥ 20-unit increase) by day 7 and day 21 are also reported for each virus (Table I). Within the IBR titer data, a greater proportion of steers in the meloxicam group seroconverted by day 21 compared to the control group (P = 0.01), resulting in a 13.82% absolute difference. However, by chance, the meloxicam group had a significantly greater proportion of steers categorized as having negative titers to BHV-1 (0 to 2 ELISA units; 53.68%) compared to the control group at arrival (40%, P = 0.03), similarly representing a 13.68% absolute difference. There were no differences regarding the proportion of steers classified as having negative titers at arrival or the proportion that seroconverted by day 7 or day 21 for BRSV, BPIV-3, or CV.

Table I.

Percent of steers that seroconverted (≥ 20 unit ELISA increase) to BHV-1, BRSV, BPIV-3 and CV by Day 7 and Day 21 by treatment group.

Virus	% Seroconverted
	Day 7	Day 21
BHV-1
Control	1.48	20.00a
Meloxicam	2.21	33.82b
BRSV
Control	1.48	84.44
Meloxicam	2.21	81.62
BPIV-3
Control	40.00	69.63
Meloxicam	36.03	64.70
CV
Control	68.89	78.52
Meloxicam	67.65	81.62

Significant differences (P < 0.05) denoted by different superscripts within a column.

Four steers were pulled and treated for BRD during the course of the study. Initial pull dates for each steer were on days 5, 10, 32, and 41. Morbidity was associated with treatment group, as all of these steers received meloxicam on arrival (control = 0.0%, meloxicam = 2.94%; Chi-square = 4.03; P < 0.05). One of these steers died (control = 0.0%, meloxicam = 0.74%; 25% case fatality rate) following 4 treatments for BRD and was diagnosed with bilateral chronic bronchopneumonia and unilateral acute fibrinous pleuritis. Small sample size precluded mortality statistical analysis.

Descriptive statistics and distribution for performance variables at each handling event by treatment group are provided (Table II). There were no differences in body weight or average daily gain at any handling events between the control and meloxicam treatment groups. Only a main effect for time (P < 0.001) existed when body weight was compared by treatment group using repeated measures ANOVA.

Table II.

Descriptive statistics for performance parameters at each handling event by treatment group.

Handling event	N	Control		Meloxicam
		Mean ± SD	Median (range)	Mean ± SD	Median (range)
Arrival
Body weight (kg)	271	296.28 ± 30.78	297.10 (212.73 to 353.80)	290.56 ± 32.02	291.43 (201.85 to 368.77)
Day 7
Body weight (kg)	270	297.89 ± 31.06	300.05 (209.11 to 366.05)	293.27 ± 33.81	293.25 (208.65 to 390.99)
ADG (kg/d)	270	0.28 ± 1.70	0.42 (−4.28 to 4.54)	0.39 ± 1.86	0.58 (−5.51 to 4.86)
Day 21
Body weight (kg)	269	322.35 ± 33.94	327.95 (226.80 to 395.53)	318.27 ± 34.22	316.61 (215.46 to 416.85)
ADG (kg/d)*	269	1.24 ± 0.57	1.17 (−0.82 to 2.87)	1.32 ± 0.59	1.32 (−2.22 to 2.76)
Day 45
Body weight (kg)	270	356.42 ± 36.38	360.61 (260.36 to 425.92)	350.52 ± 36.01	350.17 (245.85 to 434.54)
ADG (kg/d)	270	1.34 ± 0.32	1.37 (0.28 to 2.22)	1.34 ± 0.30	1.33 (0.50 to 2.19)

^*Denotes non-parametric distribution.

ADG — Average daily gain; SD — Standard deviation.

Discussion

Non-steroidal anti-inflammatories have been proposed as ancillary therapy for stressful events and painful procedures in beef cattle. However, the impact of a long-acting NSAID on primary immune response and health are not fully understood and, in some situations, appear to be detrimental. As observed by the authors in a previous in-house unpublished study, cattle receiving meloxicam experienced increased treatment risk. In this blinded clinical trial enrolling over 6500 high-risk, fall-placed calves, those receiving oral meloxicam on arrival at the feedlot were more likely to be treated for BRD throughout the feeding period (P = 0.035) (Craig Dorin, 2017, Veterinary Agri-Health Services). It was originally hypothesized that the increased treatment risk was due to decreased vaccination response, as the use of Cox-2 inhibitors in humans has been shown to negatively impact antibody production (11). Based on results in the current study, decreased vaccination response following concurrent vaccine and NSAID administration does not appear to be a factor in cattle. Alternative hypotheses that may help explain increased treatment risk among cattle receiving meloxicam include impaired inflammatory response, other immunological factors, or a combination thereof.

In regard to the perceived benefit of meloxicam as it pertains to seroconversion, it is noteworthy that the meloxicam group had a greater proportion of steers categorized as having negative BHV-1 titers upon arrival relative to the control group. Upon further examination of individual data, it became obvious that this discrepancy effectively increased the likelihood that a larger proportion in the meloxicam group would seroconvert following vaccination. This conclusion is further supported by the non-significant P-value for treatment group effect and lack of group × time interaction when evaluated using repeated measures ANOVA with the Bonferroni correction. The variation in arrival BHV-1 titers between treatment groups is viewed as the primary limitation of this study. As an aside, exceptionally potent anti-inflammatory drugs, such as glucocorticosteroids, may influence reactivation of BHV-1 from the trigeminal ganglia (3). However, this phenomenon would not be expected to hold true for a one-time administration of meloxicam at recommended dose.

As a result, the authors concluded that meloxicam had no significant biological effects on vaccine response and titers in the current study. This is in agreement with previous studies, albeit meloxicam was administered orally and to a notably smaller sample size (6,7). As such, altered vaccination response following the administration of meloxicam is likely not a justifiable concern. Other research indicates that meloxicam does not appear to have a positive effect on inflammatory mediators or leukocyte function (5,6). However, this research was performed in healthy yearling steers and may not be applicable to abruptly weaned beef calves in a commercial feedlot setting.

The lack of differences in performance parameters between treatment groups is not surprising given the overwhelming impact of compensatory gain on shrunk cattle upon entry to a feedlot. Previous research also determined that meloxicam did not appear to influence performance (7).

In summary, the association between meloxicam and morbidity and mortality events remains unclear. Perhaps there are behavioral aspects associated with NSAID administration as they relate to clinical disease that have yet to be explored. For example, meloxicam may aid in the masking of clinical signs, such as depression and pyrexia (2), typically used to justify pulling and treating an animal in the acute stages of BRD. Delayed or diminished presentation of clinical signs could subsequently allow the disease to progress to a subacute or chronic state, thereby increasing the probability of mortality. In this scenario, the onset of fatal disease would need to occur at arrival or shortly after to coincide with duration of activity of meloxicam. Incidentally, the single steer that died in the current study was initially pulled and treated on day 5. However, the 2 steers with prolonged latency to first pull do not logically fit this hypothesis.

Although this study provided valuable insight regarding the effect of meloxicam on vaccination response and treatment risk, there were some limitations. These cattle were believed to be in a higher risk category based on abrupt weaning methods and as such, were administered a macrolide antibiotic on arrival; however, their treatment rates suggest that they were likely low to medium risk. A larger group of high risk, long-hauled, auction-mart cattle may have provided the opportunity to evaluate mortality; however, it is unlikely that the calf-hood vaccination history could have been ascertained. Due to logistics, concern for cattle health and welfare, and limited funds, it was not possible to collect a 4th blood sample to establish a true peak titer response. Conversely, it is possible that titers beyond 21 d on feed could reflect exposure to circulating virus in a commercial feedlot setting as opposed to vaccination response.

In conclusion, stress associated with abrupt weaning methods and transport, coupled with the concurrent administration of modified live vaccines, long-acting NSAIDs, and metaphylactic antibiotic, likely creates a complex immunological response. Further research is warranted to explore the relationship between NSAID administration at processing and negative impacts on health, as it is likely not attributable to decreased vaccination response. In the interim, producers and veterinarians are encouraged to balance the need for mitigation of inflammation and pain with potential adverse effects on health when deciding whether to administer an NSAID such as meloxicam.