Sciatic, spinal and closely related nerve injury rodent models are in common use to study neuropathic pain. Although our experience suggests that these models may be highly sensitive to small changes in animal care, all relevant details are usually not reported in research papers, potentially resulting in both suboptimal animal welfare and increased inter-laboratory variability. While using the chronic constriction injury (CCI) model (Bennett and Xie, 1988) in Wistar rats, we observed increased rates of adverse events following a change in bedding and glove type.

Robinson et al. (2004) previously reported that paw hypersensitivity was reduced in intensity and duration in rats housed with coarse sawdust as bedding compared with those housed with soft sawdust. We have now observed that when another common type of bedding is used, corn cob (1/4” Bed-o-cobs®, The Andersons Inc., Maumee, OH, USA), the incidence of ipsilateral hindpaw oedema increased to 45%, compared with 9% while using soft (pine) wood shavings as bedding. Mild oedema was only observed in CCI rats, but not in sham surgery rats, highlighting the importance of this to nerve injury-induced pathology. Not only did the incidence of oedema increase fivefold with the change to the harder corn cob bedding, but swelling intensity also increased. We observed that in the initial stages after surgery, animals kept their ipsilateral paw in a more guarded position with the harder bedding, rarely touching the ground. We theorized that in these animals, reduced muscle movement in the ipsilateral paw resulted in lymph pooling, and therefore increased swelling in the affected foot. Corn cob bedding is now widely favoured over wood-based beddings in universities and research institutes due to increased absorbency of excreted ammonia, and reduced effects on liver enzyme function compared with wood-based beddings (Weichbrod et al., 1988). Our observations suggest that while corn cob bedding may have advantages over softer wood-based beddings in certain applications, it is probably unsuitable as a bedding substrate for rodent subjects in neuropathic pain studies.

In addition we observed that rare incidences of self-mutilation at the wound site in CCI rats, which were absent in sham-treated rats, increased in frequency following a change in glove type during surgery. Over a brief period of using natural latex gloves during CCI surgery, four out of eight animals (50%) exhibited signs of self-mutilation by gnawing at the wound site. Across a range of experiments, when using synthetic nitrile gloves only 4 out of 165 animals (2%) exhibited this behaviour. We theorized in this case that latex gloves shed allergen during surgery and exacerbated a local inflammatory response and irritation at the site of injury.

We therefore propose that (1) precise descriptions of both bedding and surgical accompaniments are included in all protocols for pain models, and neuropathic pain models in particular, (2) that soft bedding and hypo-allergenic materials be used in neuropathic pain models for both model repeatability and for animal welfare.


  1. Top of page
  2. References