The Onset of Treatment With the Antidepressant Desipramine Is Critical for the Emotional Consequences of Neuropathic Pain

Cristina Alba-Delgado; Meritxell Llorca-Torralba; Juan Antonio Mico; Esther Berrocoso


Pain. 2018;159(12):2606-2619. 

In This Article

Abstract and Introduction


Neuropathic pain is a chronic condition that is challenging to treat. It often produces considerable physical disability and emotional distress. Patients with neuropathic pain often experience depression and anxiety both of which are known to be temporally correlated with noradrenergic dysfunction in the locus coeruleus (LC) as pain becomes chronic. Antidepressants are the first-line drug therapy for neuropathic pain, and the LC represents a potential target for such therapy. In this study, we evaluated the efficacy of the tricyclic antidepressant desipramine (DMI, a noradrenaline reuptake inhibitor) in preventing or relieving the noradrenergic impairment induced by neuropathic pain. The treatment started before or after the onset of the anxiodepressive phenotype ("early or late treatment") in adult rats subjected to chronic sciatic constriction. Electrophysiological and western blotting assays showed LC dysfunction (increased bursting activity, alpha2-adrenoceptor sensitivity, tyrosine hydroxylase, and noradrenaline transporter expression) in chronic constriction injury at long term. These noradrenergic changes were concomitant to the progression of anxiety and despair-like features. Desipramine induced efficient analgesia, and it counteracted the despair-like behavior in chronic constriction injury-DMI animals, reducing the burst rate and tyrosine hydroxylase expression. Surprisingly, "early" DMI treatment did not modify pain-induced anxiety, and it dampened pain aversion, although these phenomena were abolished when the treatment commenced after noradrenaline impairment had been established. Hence, DMI seems to produce different outcomes depending when the treatment commences, indicating that the balance between the benefits and adverse effects of DMI therapy may shift as neuropathy progresses.


Epidemiological studies estimate that neuropathic pain, as a global clinical entity, affects between 0.9% and 17.9% of adult European and American people,[89] particularly women and the elderly.[19,35,48,78] Moreover, there is evidence that neuropathic pain augments the frequency and severity of emotional instability,[6,67] in particular, increasing the risk of depression and anxiety.[6,21,27,50,75,92] Indeed, approximately 20% to 30% of patients suffering chronic pain also suffer emotional-related disorders.[58,62]

The current therapeutic options to treat neuropathic pain include the use of tricyclic antidepressants.[63] These drugs relieve pain after prolonged treatment, although only mild clinical effects are achieved in many cases,[59] probably reflecting how the efficacy of these drugs in each patient depends on the effects of descending modulation on the pathophysiology of pain.[94] In addition to their analgesic effect, antidepressants improve depression and anxiety in patients with pain when these conditions are already well-established.[7,86,96] However, no controlled clinical studies have been conducted that definitively test whether antidepressant treatment administered for neuropathic pain can successfully prevent the development of comorbid emotion-related disorders. Rodent models that reproduce peripheral nerve injuries are useful tools to test and compare the effect of antidepressant drugs at different time points of neuropathic evolution.

The noradrenergic system seems to be important in regulating the sensory and emotional dimensions of chronic pain.[57] A functional impairment of the supraspinal noradrenergic system might be implicated with a worsening of mood in animals with long-term neuropathic pain.[3,17] Also, a disrupted neurochemistry of the locus coeruleus (LC) has been demonstrated in postmortem brain tissue from depressed individuals.[68,69] Such dysfunction is apparently associated with the LC and prefrontal cortex (PFC). Locus coeruleus neurons are believed to be the main source of noradrenaline in the central nervous system, and LC activity is modulated by antidepressant drugs,[49,91] and it is apparently involved in antidepressant-mediated analgesia.[4,17] As a primary noradrenergic afferent to the PFC,[26,33,65] the LC is also part of central circuitry involved in the pathophysiology of mood disorders.[5,11,23,46]

Accordingly, we postulated that early intervention with noradrenergic analgesic antidepressants might prevent the pain-induced impairment of LC function and, hence, the ensuing establishment of anxiety and despair-like behaviors. To evaluate this hypothesis, neuropathic rats subjected to chronic constriction injury (CCI) were administered desipramine (DMI), a noradrenaline reuptake inhibitor. Desipramine was chosen because it is a tricyclic antidepressant with a priority action on the noradrenergic system and with demonstrated analgesic properties in animal models and humans.[1,17,31,32,55,82] We assessed the administration of DMI in 2 conditions (Figure 1A): as an "early treatment" by treating animals before the establishment of pain-induced emotional deficits; as a "late treatment" in animals treated when the anxiety-like and depression-like features were already established, thereby analyzing the effectiveness of the treatment in restoring a normal emotional state. The sensory and emotional dimensions of pain, anxiety and despair-like behavior, and LC function were evaluated in both cases.

Figure 1.

Effect of DMI treatment on sensitivity to peripheral stimuli after CCI. (A) Representative scheme of DMI (10 mg/kg/day) treatment. Osmotic minipumps were implanted in the second (condition 1: "early treatment") or 4th week (condition 2: "late treatment") after neuropathy induction, and DMI or the vehicle (saline 0.9%) alone was delivered continuously for 2 weeks (gray area). (B–E) Sensitivity to peripheral stimuli of the ipsilateral hind paw: (B) withdrawal threshold (g) in response to von Frey hair stimulation; (C) number of hind paw lifts in the cold plate test; (D) withdrawal latency (s) to heat stimulus; (E) withdrawal threshold (g) in response to paw-pressure stimulation. Symbols represent the mean ± SEM of (n) animals per group. Sham-control and CCI-control groups represent pooled saline-treated and untreated animals: **P < 0.01 and ***P < 0.001 vs baseline (B); +P ≤ 0.05, ++P < 0.01, and +++P < 0.001 vs CCI group (repeated-measures 2-way ANOVA followed by Tukey–Kramer HSD post-tests). ANOVA, analysis of variance; CCI, chronic constriction injury; DMI, desipramine; HSD, honestly significant difference.