Research on the effects of opioids on immune responses was stimulated in the 1980s by the intersection of use of intravenous heroin and HIV infection, to determine if opioids were enhancing HIV progression. The majority of experiments administering opioid alkaloids (morphine and heroin) in vivo, or adding these drugs to cell cultures in vitro, showed that they were immunosuppressive. Immunosuppression was reported as down-regulation: of Natural Killer cell activity; of responses of T and B cells to mitogens; of antibody formation in vivo and in vitro; of depression of phagocytic and microbicidal activity of neutrophils and macrophages; of cytokine and chemokine production by macrophages, microglia, and astrocytes; by sensitization to various infections using animal models; and by enhanced replication of HIV in vitro. The specificity of the receptor involved in the immunosuppression was shown to be the mu opioid receptor (MOR) by using pharmacological antagonists and mice genetically deficient in MOR. Beginning with a paper published in 2005, evidence was presented that morphine is immune-stimulating via binding to MD2, a molecule associated with Toll-like Receptor 4 (TLR4), the receptor for bacterial lipopolysaccharide (LPS). This concept was pursued to implicate inflammation as a mechanism for the psychoactive effects of the opioid. This review considers the validity of this hypothesis and concludes that it is hard to sustain. The experiments demonstrating immunosuppression were carried out in vivo in rodent strains with normal levels of TLR4, or involved use of cells taken from animals that were wild-type for expression of TLR4. Since engagement of TLR4 is universally accepted to result in immune activation by up-regulation of NF-κB, if morphine were binding to TLR4, it would be predicted that opioids would have been found to be pro-inflammatory, which they were not. Further, morphine is immunosuppressive in mice with a defective TLR4 receptor. Morphine and morphine withdrawal have been shown to permit leakage of Gram-negative bacteria and LPS from the intestinal lumen. LPS is the major ligand for TLR4. It is proposed that an occult variable in experiments where morphine is being proposed to activate TLR4 is actually underlying sepsis induced by the opioid.