Sex hormones and pain

The three main sex hormones (i.e. oestrogen, progesterone and testosterone) are functionally active in both sexes, but their absolute levels and temporal fluctuations differ considerably in males and females (Berkley 1997, Berkley & Holdcroft 1999, Cairns & Gazerani 2009). Females undergo vast hormonal changes during puberty, pregnancy and menopause and cyclic hormonal fluctuations during the ovarian cycle in the reproductive phase of life. Males are instead exposed to less marked fluctuations in hormone levels across the lifespan, with the most significant change being the reduction of testosterone with ageing (Fillingim et al. 2009). Among the many body function parameters influenced by sex hormones, pain perception holds an important place, although there is not always universal agreement about how and to what extent this happens throughout the lifespan nor about the pathophysiology of these differences (Giamberardino 2000). In women, many painful conditions vary in their incidence, disappearance and prevalence as a function of puberty, pregnancy, menopause and ageing and, as already mentioned, during the reproductive years different forms of pain also vary with the phase of the menstrual cycle, mostly exacerbating in the perimenstrual period. In men, some pain disorders also show different profiles in the various stages of life (Berkley 2005, Cairns & Gazerani 2009, Fillingim et al. 2009).

Progesterone is mostly associated with analgesia because some pain conditions in humans – such as migraine and temporomandibular pain – disappear or improve during pregnancy or the midluteal phase of the menstrual cycle, and other pains are reduced in animals during lactation (when progesterone levels are high), and some anaesthetics are progesterone-based (e.g. alphaxolone) (Berkley & Holdcroft 1999, Silberstein 2004, LeResche et al. 2005, Brandes 2006, Craft 2007). Oestrogen has also been associated with analgesia, since some pain conditions increase when oestrogen decreases. For instance, as the oestradiol level sharply declines postpartum, the frequency of migraine attacks increases (Sances et al. 2003) and after the menopause, when oestrogen declines, several pain complaints – such as orofacial pain and vaginal pain – increase (LeResche et al. 2003, Fillingim et al. 2009). Similarly, testosterone promotes analgesia, its decline with ageing in men being consistently associated with an increase in a number of pains, such as angina or muscle pain (Berkley & Holdcroft 1999, Vecchiet 2002). For each of these examples, however, either a lack of effects or contrasting examples can be found, such as the decrease in postmenopausal women of musculoskeletal pain, chronic widespread pain and fibromyalgia, and in postmenopausal women and older men of abdominal pain (including irritable bowel syndrome, IBS) migraine and tension headaches, in parallel with a decrease in oestrogen, progesterone and testosterone. Another example is the emergence of cluster headaches in men at puberty, when testosterone increases (Berkley & Holdcroft 1999, LeResche et al. 2003, Kuba & Quinones-Jenab 2005). Exogenous hormone use has also been associated with change in several pain patterns. Women under oral contraceptive treatment have an increased risk for development of temporomandibular (TMD) pain and carpal tunnel syndrome.

Postmenopausal women under hormone replacement are at increased risk of back pain and TMD pain (LeResche et al. 1997, Ferry et al. 2000, Musgrave et al. 2001), but also discontinuation of this therapy is associated with higher levels of reported pain or stiffness (Ockene et al. 2005). Likewise, after sustained oestradiol administration, migraine attacks are precipitated by oestradiol withdrawal (Lichten et al. 1996). An interesting study in transsexuals taking hormones to acquire characteristics of the opposite sex has shown changes in pain responses, with over 30% of those taking oestradiol/antiandrogen developing chronic pain and 50% of those taking testosterone reporting improvement of chronic pain (headache) present before start of treatment (Aloisi et al. 2007). Thus both administration and withdrawal of exogenous oestrogens – but not testosterone – appear associated with an increased risk of chronic pain.

All these apparent contradictions can, at least partially, be accounted for by the fact that the overall hormonal effects on clinical pain expression depend more on the concentration of one hormone relative to the others than on its absolute values (Fillingim et al. 2009; see also Giamberardino 2000). The modulation of pain perception by sex hormones is probably the result of a combination of factors, among which the hormonal influences on metabolism (with implications for drug action), the immune system (with implications for painful autoimmune diseases, up to nine times more common in women), trauma-induced inflammation (modulated by sex hormones), the hypothalamic–pituitary axis (with implications for the interactions between stress and pain), and nervous and cardiovascular systems (Fillingim et al. 2002, al’Absi et al. 2004, Aloisi & Bonifazi 2006, Craft 2007, Straub 2007).

As already discussed, however, sex hormone effects on pain perception cannot be separated from the many other variables that affect pain modulation. Of particular importance are social and cultural factors, which can entail profound diversities in men and women (both patients and physicians) in their attitude towards and approach to painful symptoms, especially regarding particular forms of pain such as those arising from the pelvic area (see below) (Myers et al. 2003).

Visceral pain

The clinical presentation of pain from internal organs typically varies with time. In the first phases of a visceral algogenic process the symptom is very aspecific, always perceived in the same site – along the midline of the thorax or abdomen – whatever the viscus in question. It is vague and poorly discriminated, often described more as a sense of oppression or malaise rather than frank pain, and is accompanied by marked neurovegetative signs and emotional reactions (true visceral pain) (Procacci et al. 1986, Giamberardino 2005). In a second phase – after a few minutes or a few hours in the first episode or in a subsequent episode – it becomes referred to somatic structures of the body wall, in areas that are neuromerically connected to the viscus in question. Examples are the left chest area and ipsilateral upper limb for the heart, the lumbar region–flank–anterior abdomen spreading to the groin for renal colics, the upper right abdominal quadrant radiating towards the back at the inferior angle of the scapula for biliary colics, or the lowest abdominal quadrants and sacral region for pain from the female reproductive organs. In this phase, the symptom is very similar to that originating directly from the somatic structures; its visceral origin can thus be difficult to identify. In the referred pain area, hyperalgesia (i.e. an increased sensitivity to painful stimuli) typically develops. This is mostly localized in the skeletal muscle layer, but during particularly prolonged and/or intense processes it spreads upward to also involve the overlying superficial somatic tissues – the subcutis first, and finally the skin. Cutaneous hypersensitivity can become frank allodynia (pain perceived for normally non-painful stimuli) in extreme cases, as happens in peritonitis from painful abdominal conditions, such as a perforated appendicitis. Vice versa, in the course of the healing process of the painful visceral condition, the desensitization of the somatic area of referral proceeds from the surface downwards; the skin is the first to normalize and then the subcutis, with the muscle keeping some degree of residual hyperalgesia for a very long period of time (Giamberardino & Cervero 2007). The characteristics and temporal evolution of the referred hyperalgesia have been deducted from a number of studies in patients with visceral pains of various origin, such as urinary/biliary colics (from calculosis or diskinesia), IBS, dysmenorrhea, endometriosis. Most of this research has employed a combination of both clinical and instrumental procedures to assess the hypersensitivity, the latter involving measurement of pain thresholds to various stimuli (thermal, mechanical, chemical and electrical). Hyperalgesia is demonstrated by a significant lowering in pain detection threshold (Vecchiet et al. 1989, 1992, Giamberardino et al. 1994, 2001, 2005, Caldarella et al. 2006).

The global outcome of the studies performed indicates that hyperalgesia only appears in visceral conditions that are painful, irrespective of the nature of the visceral trigger (organic or dysfunctional), but is absent in any organic condition that is not algogenic. As already mentioned, the hyperalgesia is mostly a muscle phenomenon, involving the overlying subcutis and skin tissues only in more severe cases. Muscle hyperalgesia occurs early in the course of the visceral algogenic process (i.e. a few painful episodes are sufficient for it to manifest), and is accentuated by repetition of the episodes as the degree of pain threshold lowering becomes progressively more pronounced. Furthermore, it is of long duration; it normally outlasts the spontaneous pain – being detectable in the pain-free interval – and in most cases even the primary insult in the internal organ, though in a milder form. For instance, the majority of patients with urinary colics from calculosis who have passed the stone still present some degree of referred muscle hyperalgesia in the lumbar region months or even years after elimination (Vecchiet et al. 1992, Giamberardino et al. 1994).

Referred hyperalgesia is usually accompanied by trophic changes of local deep somatic wall tissues, mostly consisting of increased thickness of the subcutis and decreased thickness/section area of muscle, the latter testifying a tendency towards atrophy of muscle layers. These can be documented by clinical means (i.e. pinch palpation) but better quantified by ultrasonography. In symptomatic urinary and biliary calculosis, in fact, a significant increase in subcutis thickness and a significant decrease in muscle thickness have been found in the referred area (lumbar region and cystic point area, respectively) with respect to the contralateral non-affected area (Giamberardino et al. 2005, Giamberardino & Cervero 2007). Like the hyperalgesia, also trophic changes are set off by the algogenic impulses from the affected organ, since they are not detected in non-painful organic visceral conditions such as asymptomatic gallbladder calculosis. Unlike the hyperalgesia, however, they are not modulated by the extent of algogenic impulses from the visceral organ. In fact, they have been shown not to increase with the repetition of the painful episodes or decrease with their cessation; they seem to be a rather on–off phenomenon (Giamberardino 2005).

Referred pain with hyperalgesia has been attributed to phenomena of central sensitization involving viscerosomatic convergent neurons (Woolf & Salter 2000). The afferent barrage from the affected organ would increase the activity and response properties of these neurons, thus enhancing the central effect of the normal input from the somatic area of pain referral and accounting for the hyperalgesia (convergence-facilitation theory) (Cervero & Laird 2004, Sengupta 2009). (See discussion of these issues in Chapter 3.)

The visceral input would also activate a number of reflex arcs, whose afferent branch is represented by sensory fibres from the organ and whose efferent branch would be somatic towards the skeletal muscle and sympathetic towards the subcutis and skin of the referred area. Activation of these reflexes would contribute to the secondary hyperalgesia and also account for the local trophic changes (see Giamberardino et al. 2005). Hyperalgesia and trophic changes can be typically detected in referred pain areas from pelvic internal organs, as will be reported in the following sections.

Pain from the female reproductive organs