Central Sensitization Inventory: What Your Score Can and Cannot Say

A replaced knee can look perfect on an X-ray and still hurt, and the proof of why often sits in the forearm. When skin with nothing wrong with it hurts at half the pressure a healthy person tolerates, the problem has left the joint. Many patients meet that problem first as a number: a score on the central sensitization inventory, handed back with no explanation of what it means.

The video above, Chapter 9 of The Pained Brain, explains how the nervous system becomes an amplifier. Below: how the amplifier is measured, how far to trust each tool, and what a positive result predicts. The book is by Dr. Gurpreet Singh Padda, MD, MBA, MHP, and Dr. KrisJay Fucanan, MD.

The amplifier, in one cascade

Central sensitization is a lasting but reversible rise in how excitable the pain neurons of the spinal cord and brain become. Lasting means it outlives the injury. Reversible means it is not damage. Nothing is broken. Something has been learned.

The learning follows a chain. An inflamed joint keeps firing into the cord. The NMDA receptor on the receiving neuron, normally plugged by a magnesium ion, opens. The synapse strengthens the way a practiced skill strengthens. Then activated microglia, the resident immune cells of the nervous system, release a growth factor that disables the chloride pump the brakes depend on, so the inhibitory signal stops inhibiting. Inputs that never counted as pain now do. Light touch hurts. A pinprick repeated once a second climbs with every tap.

How strongly a nervous system amplifies is a personal trait. In fifty healthy men sensitized on four separate days, a man’s own result barely moved, a correlation of 0.85, while the correlation between different men was 0.03.

What the central sensitization inventory measures

The inventory is a symptom questionnaire. The classic cutoff is 40. In 2,890 patients filling it out before a first visit at one Dutch pain center, the average score sat right at that line, at 40.2 against 16.4 in healthy volunteers, and 59.9 percent reported at least one central sensitivity syndrome.

The same data suggest the classic line is too high. The value that best separated patients from healthy controls was 30, with an area under the curve of 0.95, and it was 33 in women and 25 in men. That comes from one center and a Dutch-language version, and its authors warn against carrying the cutoffs elsewhere. Read a borderline score with that in mind.

Now the limit. Across 66 studies and 13,284 people, inventory scores tracked depression, anxiety and catastrophizing strongly, and laboratory measures of nerve sensitivity weakly. So the score is a screen for the syndrome. It is not a measurement of the synapse. A high score is a reason to measure, never a verdict about your character.

The measurement: pressure, wind-up and the brakes

Three bedside tests look at the physiology directly. A pressure algometer, a calibrated gauge, is pressed on a site far from the pain until you first say pain. A stimulus repeated once a second shows temporal summation, the spinal cord’s wind-up. Conditioned pain modulation applies one pain while testing another, to see whether your body’s own descending brakes engage.

• Pressure threshold: reliability of 0.77 or better across 88 people with rheumatoid arthritis, back pain or no pain, tested by different raters one to three weeks apart.
• Temporal summation: 0.76 or better in the same study, and scored better as a difference than a ratio.
• Conditioned pain modulation: the noisy one. Its best within-session reliability was 0.64 in healthy people and 0.77 in patients, and 24 of 25 studies were rated inadequate on reporting.

One abnormal reading is a research measurement, so the test gets repeated. The point of an instrument is the second reading.

Why the blood test missed it

In 160 people, 40 with degenerative disc pain and 40 with disc herniation plus two control groups, 34 inflammatory proteins ran high in the spinal fluid of each patient group. In the same patients’ blood, the inflammatory proteins ran below the controls. A normal CRP does not mean nothing is inflamed. It means the blood was the wrong place to look.

The drivers are metabolic, and at least two are measurable. In 265 adults with knee pain, low pressure thresholds went with more visceral fat, a median 130 against 95 square centimeters, and more leptin, 22.2 against 13.3 ng/mL. Both held in people whose BMI was normal: thin outside, inflamed inside. CRP, fasting glucose and HbA1c showed no association at all. And in 302 healthy, pain-free adults, a heavier cardiometabolic load already went with more spinal facilitation and weaker inhibition, before any of them had a pain condition. The gut is part of that terrain, which is why the gut’s reach into the spine belongs in the same conversation.

The third driver is social. Being disbelieved is a stressor, and chronic stress lowers the pain threshold. Invalidation by one’s own family raised the odds of the fibromyalgia phenotype 1.81-fold, separate from depression.

What sensitization predicts before a procedure

This is why I test before I schedule. Among 103 knee replacement patients sorted before surgery, those with both facilitated wind-up and failed brakes got 52.0 percent relief at a year, against 81.1 and 79.6 percent in those with only one abnormality. In 464 hip and knee replacements, each point on a widespread-pain score raised the odds of failing to halve the pain by 17.8 percent, and failure ran 26.97 percent in the lowest third against 60.68 percent in the highest. In 37 patients over 65 given genicular radiofrequency plus a joint injection, a score of 40 or more blunted the pain response by 2.1 points.

The opposite finding matters as much. After discectomy for a herniated disc, 100 patients improved whatever their score, and the average score itself fell from 22.3 to 13.7. Where there is a true mechanical lesion, fix it. The test does not cancel a procedure. It tells me whether the plan for the months after the procedure needs to be ready before the needle goes in, which also explains why each injection can buy less time than the last.

Turning the amplifier down

The amplifier is maintained by input. In 53 people with chronic whiplash, radiofrequency neurotomy of the neck joint nerves reduced body-wide pressure and heat hypersensitivity at one and three months. When the nerves regrew and the pain returned at about ten months, the central signs came back with it. That is the bridge seen from the spinal cord: a block or an ablation does not cure the amplifier, it stops feeding it.

The silence has to be used. In 110 adults who lost 7.9 percent of their weight by diet over three months, chronic musculoskeletal pain fell from 51 percent to 25 percent, with no change in hsCRP. It was uncontrolled and short, the weakest design in the literature and the largest effect in it. Food, sleep and movement are prescribed like a procedure, because they remove the fuel, and exercise is dosed like a drug, since the pain relief a healthy person gets from a workout is not reliably present in a sensitized one. Fear is treated too, with Acceptance and Commitment Therapy delivered in-house.

The drugs aimed at the amplifier are modest: pregabalin in fibromyalgia has a number needed to treat of 7 to 14. Any medication change belongs with your own physician. The humility here is mine: my specialty built its tools for the joint, and the problem moved into the cord.

Questions to bring to your appointment

• Can an algometer be pressed on a part of me that does not hurt, and repeated?
• Before this procedure, has anyone checked my wind-up and my brakes?
• Can we measure visceral fat and leptin, not only weight?
• What is the plan for the months a procedure buys?

The Technical Supplement to Chapter 9 sets out every study cited here, the number with its population, and what each one does and does not show, written to be handed to your doctor. When the amplifier drives someone to the emergency room at two in the morning, what that room can and cannot find is the next chapter.

Frequently asked questions

Is central sensitization real, or is the pain in my head?

It is a physical state of the nervous system. PET tracers have shown activated glia in the brains of people with chronic back pain, fibromyalgia and knee arthritis, and spinal fluid in disc patients carries elevated inflammatory proteins. The pain is produced by the brain and cord, as all pain is, but the amplification can be measured with an algometer. Here is why a scan so often fails to match the pain you feel.

What is a high score on the central sensitization inventory?

The classic cutoff is 40. In one Dutch pain center, a lower value of 30 best separated patients from healthy volunteers, with 33 for women and 25 for men. A high score shows that the pattern of symptoms fits sensitization, but the questionnaire also tracks mood and hypervigilance, so it should prompt direct testing rather than end the discussion. See why normal test results are expected in fibromyalgia.

Why does my knee still hurt after a successful replacement?

About one in five knee replacements leaves chronic pain even when the implant is perfect. Patients with both facilitated wind-up and failed descending brakes before surgery got far less relief than those with one abnormality. When the pain has moved into the nervous system, a new joint does not switch it off, and the next step is to find the real driver. Read how persistent pain after knee replacement is worked up.

Can a blood test show central sensitization?

Not reliably. In disc patients, inflammatory proteins were high in spinal fluid while the same proteins in blood ran below controls. In adults with knee pain, CRP, fasting glucose and HbA1c were not linked to low pain thresholds, while visceral fat and leptin were. Measuring the fat that signals, rather than the marker that circulates, tells you more. Learn how leptin resistance runs pain hotter than scans suggest.

Can central sensitization be reversed?

Often, yes. After knee replacement, cervical radiofrequency and hip replacement, widespread hypersensitivity eased once the painful input stopped, and in whiplash it returned when the pain came back. In a minority it persists. Removing the input buys a window, and the metabolic, sleep and movement work done inside that window is what keeps the volume down. Pain is the signal, not the disease, and that changes the treatment.

Sources

1. Woolf, C. J. (2011). Central sensitization: implications for the diagnosis and treatment of pain. Pain, 152(3 Suppl), S2-S15. https://doi.org/10.1016/j.pain.2010.09.030
2. Schuttert, I., Wolff, A. P., Schiphorst Preuper, R. H. R., Malmberg, A. G. G. A., Reneman, M. F., & Timmerman, H. (2023). Validity of the Central Sensitization Inventory to Address Human Assumed Central Sensitization: Newly Proposed Clinically Relevant Values and Associations. Journal of Clinical Medicine, 12(14), 4849. https://doi.org/10.3390/jcm12144849
3. Adams, G. R., Gandhi, W., Harrison, R., van Reekum, C. M., Wood-Anderson, D., Gilron, I., & Salomons, T. V. (2023). Do “central sensitization” questionnaires reflect measures of nociceptive sensitization or psychological constructs? A systematic review and meta-analyses. Pain, 164(6), 1222–1239. https://doi.org/10.1097/j.pain.0000000000002830
4. Brady, S. M., Georgopoulos, V., Veldhuijzen van Zanten, J. J. C. S., Duda, J. L., Metsios, G. S., Kitas, G. D., Fenton, S. A. M., Walsh, D. A., & McWilliams, D. F. (2023). The interrater and test-retest reliability of 3 modalities of quantitative sensory testing in healthy adults and people with chronic low back pain or rheumatoid arthritis. Pain Reports, 8(6), e1102. https://doi.org/10.1097/PR9.0000000000001102
5. Rosenström, A. H. C., Ahmed, A. S., Kultima, K., Freyhult, E., Berg, S., Bersellini Farinotti, A., Palada, V., Svensson, C. I., & Kosek, E. (2024). Unraveling the neuroimmune interface in chronic pain-the association between cytokines in the cerebrospinal fluid and pain in patients with lumbar disk herniation or degenerative disk disease. Pain, 165(7), e65–e79. https://doi.org/10.1097/j.pain.0000000000003175
6. Andersson, M. L. E., Thorén, E., Sylwander, C., & Bergman, S. (2023). Associations between chronic widespread pain, pressure pain thresholds, leptin, and metabolic factors in individuals with knee pain. BMC Musculoskeletal Disorders, 24(1), 639. https://doi.org/10.1186/s12891-023-06773-4
7. Petersen, K. K., Graven-Nielsen, T., Simonsen, O., Laursen, M. B., & Arendt-Nielsen, L. (2016). Preoperative pain mechanisms assessed by cuff algometry are associated with chronic postoperative pain relief after total knee replacement. Pain, 157(7), 1400–1406. https://doi.org/10.1097/j.pain.0000000000000531
8. Brummett, C. M., Urquhart, A. G., Hassett, A. L., Tsodikov, A., Hallstrom, B. R., Wood, N. I., Williams, D. A., & Clauw, D. J. (2015). Characteristics of fibromyalgia independently predict poorer long-term analgesic outcomes following total knee and hip arthroplasty. Arthritis & Rheumatology, 67(5), 1386–1394. https://doi.org/10.1002/art.39051
9. Smith, A. D., Jull, G., Schneider, G., Frizzell, B., Hooper, R. A., & Sterling, M. (2014). Cervical radiofrequency neurotomy reduces central hyperexcitability and improves neck movement in individuals with chronic whiplash. Pain Medicine, 15(1), 128–141. https://doi.org/10.1111/pme.12262
10. Ward, S. J., Coates, A. M., Carter, S., Baldock, K. L., Berryman, C., Stanton, T. R., Yandell, C., Buckley, J. D., Tan, S.-Y., Rogers, G. B., & Hill, A. M. (2024). Effects of weight loss through dietary intervention on pain characteristics, functional mobility, and inflammation in adults with elevated adiposity. Frontiers in Nutrition, 11, 1274356. https://doi.org/10.3389/fnut.2024.1274356

Medically reviewed by Dr. Gurpreet Singh Padda, MD, MBA, MHP, board certified in anesthesiology, pain medicine, interventional pain management, addiction medicine and obesity medicine. Adapted from Chapter 9 of The Pained Brain by Dr. Padda and Dr. KrisJay Fucanan, MD. Last reviewed September 2026.

Dr. Gurpreet Singh Padda, MD, MBA, MHP