Symptoms of phantom limb pain are highly individualized, may be extremely traumatic

Mccaffrey and Beebe provide a reminder that is especially important for treating phantom limb pain victims: "Pain is whatever the experiencing person says it is, existing whenever the experiencing person say it does". A healthcare provider cannot know another person's subjective experience and exhibiting this kind of audacity directly increases patient's suffering. In a survey of 2700 veterans who had undergone amputation, 60% "reported that their physicians had directly stated or clearly implied that their pain was 'just in their heads'". (Davis, 1993, 80) Furthermore, the majority of patients were concerned that disclosing their pain to their physicians would lead them to believe they were mentally ill. Even though these figures are outdated, this attitude may still prevail today with some health care providers, and it's important to respect the patient's experience to ensure people get the care and treatment they deserve. 

A variety of different pain symptoms may be felt

http://www.luggageonline.com/lolnews/stephen-sumner-using-mirror-therapy-treat-phantom-pain/

Pain varies in frequency, intensity, duration, and type. Sometimes pain is infrequent and sharp, whereas dull, continuous pain has been noted by others. Descriptors range from "sharp, shooting, or electrical like to [...] dull, squeezing, and cramping." (Hsu et. al, 2013)  The pain may be localized to one part of the missing limb or may include the whole missing limb. Onset usually occurs soon after the amputation. Sometimes PLP increases over time, and other times it decreases. As noted in an earlier entry, factors such as age, gender, ethnicity and health status have no correlation with the existence of pain. It is important to note that PLP occurs in congenital limb loss, so nerve damage is not necessarily a factor. 

Sources: 

Davis, R. W. (1993, January). Phantom Sensation, Phantom Pain, and Stump Pain. Arch Phys Med Rehabil, 74, 79-91.

Hsu, E., & Cohen, S. P. (2013, February 13). Postamputation pain: epidemiology, mechanisms, and treatment. Journal of Pain Research, 6, 121-136.

Diagnosis of PLP is not as obvious as one would think

Diagnosis of PLP seems obvious, right? Somebody feels agonizing pain in their lost limb, and of course we know what it is. Well, sometimes, but other conditions must be ruled out to ensure the treatment fits the problem.

Conditions to Rule Out

Psychogenic limb pain--very rare and impacted by emotional and mental problems of the patient. It occurs when "the patient misinterprets [sic] nonpainful phantom sensations" (Jain, 2011).

Residual Limb Pain (RLP)--Although it triggers phantom pain, it is localized in the stump and not associated with central mechanisms. A differentiation method showing that PLP is diminished only by IV morphine but RLP is diminished by both IV morphine and lidocaine is supported by "A randomized, double-blind, active placebo controlled, crossover trial" (Jain, 2011).

Lesions of the brachial plexus and spinal nerve roots--If amputees have been pain free in the past, this is a potential hypothesis related to local nerve damage unrelated to amputation and occurring at a later date.

Lesions of the central nervous system--stroke or spinal cord injury can cause neuropathic pain in the phantom limb, yet the quality of pain may help differentiate.

Phantom Sensations (PS)--this may not be possible to differentiate because the some of the same mechanisms as PLP are implicated. Furthermore, patients are frequently unable to distinguish between the two, such as when a contorted phantom limb sensation causes pain.

Diagnostic Workup

The optimal diagnostic workup for PLP includes a thorough evaluation preceding amputation, including pain and psychological assessments. Following amputation, a pain assessment and neurological examination should be performed. Tinel sign and and MRI can detect the presence of a neuroma, which is hypothesized to be one component of PLP (and also RLP), so that could contribute to confusion.

No online videos of Tinel Sign in amputees, but this gives you an idea.

https://www.youtube.com/watch?v=RV2Ic7sY9M4

Sources:

Jain, K. (2011). Phantom Limb Pain Diagnostic Workup. In MedMerits Free access to information doctors rely on. Retrieved March 26, 2015, from http://www.medmerits.com/index.php/article/phantom_limb_pain/P8

Jain, K. (2011). Phantom Limb Pain Differential Diagnosis. In MedMerits Free access to information doctors rely on. Retrieved March 26, 2015, from http://www.medmerits.com/index.php/article/phantom_limb_pain/P7

Clarke, C., Lindsay, D. R., Pyati, S., & Buchheit, T. (2013, June). Residual Limb Pain is Not a Diagnosis A proposed Algorithm to Classify Postamputation Pain. Clin J Pain, 29(6), 551-562.

The cause of PLP remains a mystery, but several factors may contribute at once

Factors such as gender, age, location, and level of amputation, and mechanism of amputation (surgical vs. traumatic) show no correlation to incidence rates of PLP. Even though about most phantom pain begins in the first 24 hours after limb removal, it has started decades later for other patients. Usually, phantom pain is not continuous, but rather episodic, occurring for seconds to hours at a time. Sometimes it goes away for good.

Major research fails to address morbidity from PLP, yet one can imagine the debilitating effect of agonizing pain. Mortality directly caused by PLP is difficult to imagine.

Central and/ or peripheral nervous system mechanisms have been proposed for PLP, yet the exact cause still remains a mystery. To (hopefully) keep you from nodding off, I'll limit my discussion of etiology and pathophysiology to the strongest theory, proposed by Ramachandran and Hirstein. This "multifactorial model" cites five contributing factors, taking into account the considerable variation among individual experiences. 

1. Residual limb neuromas: once a nerve is severed, it may sprout branches that lead to increased sensitivity and inappropriate ie. nonprotective input to the brain. This is a peripheral factor.

2. Cortical remapping: magnetoencephalography has provided evidence that "cortical areas representing the amputated extremity are taken over by neighboring representational zones in both the primary somatosensory cortex and motor cortex." (Weeks et. al, 2010, 279) Since this is one of the most accepted theories behind PLP, and it requires a little background, I've placed a fabulous Khan academy video below that gives a little background into the somatosensory cortex (which is conceptually similar to the motor cortex).

Khan Academy: how the somatosensory cortex works

https://www.youtube.com/watch?v=3jf2l9ma6SM

I have to confess, I haven't been able to find a source that explains exactly why cortical remapping causes pain. 

I have to confess that I don't understand quite why cortical remapping causes pain, and I have been able to find a source with a clear explanation, so I hope to get back to this. Note that this is a central problem. Below I've included a picture illustrating cortical remapping, which explains the what but not the why.

How cortical remapping

http://www.google.com/patents/US20110065505

3. Monitoring of corollary discharge from motor commands to the limb. It's explained like this: "Rather than the brain receiving information that the limb is immobile, it fails to receive feedback from a newly amputated limb to confirm that a motor command has been followed." (Weeks et. al., 2010, 280) Another central problem.

4. One's own body image. This is the most self-explanatory central problem.

5. "Vivid somatic memories of painful sensations or posture of the original limb being 'carried' over into the phantom." (Weeks et. al, 2010, 280)

EDIT: 

This TED Talk by Dr. Ramachandran added clarity to the idea of cortical remapping. 

https://www.youtube.com/watch?v=l80zgw07W4Y

Sources:

Weeks, S. R., Anderson-Barnes, V. C., & Tsao, J. W. (2010, September). Phantom Limb Pain Theories and Therapies. The Neurologist, 16(5), 277-286.

Canavero, S. (1994). Dynamic Reverberation. A Unified Mechanism for Central and Phantom Pain. Medical Hypotheses, 42, 203-207.