Case Report

Treatment induced neuropathy of diabetes (TIND)

Summary

A 20-year-old male patient was admitted to intensive care unit (ITU) with severe diabetic ketoacidosis (DKA). He was treated in accordance with the Joint British Diabetes Society (JBDS) DKA guidelines. Subsequently, he was diagnosed with new onset Type 1 Diabetes (T1DM). On discharge, he was managed with subcutaneous insulin injections. His glycated haemoglobin (HbA1c) decreased rapidly from >130 mmol/mol (>14%) at diagnosis to 76 mmol/mol (9.1%) within 6 weeks of initiation of insulin therapy. This rapid and significant decline in HbA1c has consequently resulted in “treatment induced neuropathy of diabetes (TIND)” or “Insulin neuritis”. He presented with disabling painful peripheral neuropathy of distal lower limbs, severely impairing his mobility. Symptoms gradually improved after being treated with a combination of duloxetine and gabapentin for more than 6 months. Meanwhile efforts were made to stablise his glycaemic control without further intensification. This case highlights that sudden improvement in glycaemic control in patients with long standing hyperglycaemia can result in disabling iatrogenic sensory neuropathy.

Case presentation

A 20-year-old male student was brought to A&E with difficulty breathing and low GCS. He did not have significant past medical history. Prior to admission, he had not been feeling well for few days with viral symptoms. Fellow students last saw him awake at 3 am on the day of admission and found him unconscious and struggling to breathe 12 hours later.

On arrival, his GCS was 3/15. He had hypothermia (tympanic temperature 28C), profoundly low blood pressure, shallow breathing and unrecordable oxygen saturation on pulse oximetry. On examination, heart sounds were normal. Respiration was very shallow with a respiratory rate of 8/min. Chest wall movements were equal but had very quiet breath sounds. There was no signs of aspiration. Pulse was 75/min, with regular rhythm on ECG. Systolic blood pressure was only 47 mmHg. Pupils were 3mm, equal and reactive to light. As he was critically unwell, the intensivist immediately attended him in A&E and transferred to ITU: airway maintained, arterial line inserted, fluid resuscitated, and body temperature warmed up gradually.

Venous blood gas (VBG) revealed severe metabolic acidosis with pH 6.85, HCO3 6.6, pCO2 4.67, and lactate 3.8. Other laboratory results showed raised white cell count (18.86), neutrophilia (13.01), and normal inflammatory marker (CRP 3 mg/L). Renal function was also impaired with high potassium (6.6 mmol/l), urea (16 mmol/l) and creatinine (209 µmol/l) and low eGFR (38 ml/min/1.73m2). Initially glucose measured on capillary glucose was unrecordably high. Laboratory sample later confirmed it as 42 mmol/l. Plasma ketone was also raised (6.6 mmo/l). Liver function and clotting profile were normal. Other possible causes of metabolic acidosis were also excluded. Paracetamol, salicylate, and ethanol level were all within normal range. There was no clinical signs of infection. CXR did not show any consolidation or pneumothorax. Viral PCR was negative for both COVID and RSV. There was no growth on blood culture.

He was managed as a case of severe diabetic ketoacidosis (DKA) and treated immediately with fixed rate insulin infusion (FRII), intravenous fluids, prophylactic LMWH and empirical broad-spectrum antibiotics. Hyperkalaemia was managed with calcium gluconate infusion.

HbA1c at the time was >130 mmo/mol (>14%). GAD antibody, Islet antigen 2 antibody, and Zinc transporter antibody were strongly positive whereas Islet cell antibody was negative. Serum C-peptide was low, < 303 pmol/l (normal range 256-1718) indicating insulin deficiency. Subsequently, he was diagnosed with Type 1 Diabetes (T1DM). Patient was educated on self-management of diabetes, glucose and ketone monitoring by finger prick testing and insulin administration. He was also given the interstitial continuous glucose monitoring device, Libre2 sensors, so that he can scan the glucose readings with the smart phone anytime anywhere for conveniece. After 5 days of hospital admission, he was successfully discharged, on subcutaneous multiple daily insulin (MDI) regimen: Tresiba 10 units once a day as background insulin and Novorapid 4-10 units as rapid acting insulin with meals. He was also reviewed by the diabetes multidisciplinary team (MDT) during admission and was arranged to be followed up in diabetes clinic post-discharge.

Patient was very conscientious and compliant with diabetes management. He took education onboard, learned carbohydrate counting very quickly and started taking Novorapid 1 unit for 10 grams of carbohydrate after seeing dieticians. He also titrated background insulin gradually and by the time he was reviewed again in diabetes clinic 6 weeks later, he was taking 22 units of Tresiba, and had 46% of the glucose readings within target. His HbA1c dropped to 76 mmol/mol (9.1%). Further plan was for him to attend structured education programme for people with T1DM, Dose Adjustment For Normal Eating (DAFNE).

Two weeks later, he was seen urgently in the Diabetes Drop-in Clinic. He presented with severe burning sensation in both sole and feet bilaterally ascending up to the ankle level. He was unable to bear weigh and struggling to walk due to allodynia. Occasionally, pain woke him up from sleep. He could no longer play football or rugby. He had to use crutches or wheel chair at the pinnacle of presentation. On examination, there was no muscle wasting or weakness. Temperature, vibration and pinprick sensation were lost up to the ankle. Reflexes were absent bilaterally at the Achilles tendon. Regular analgesia such as paracetamol and codeine did not have any effect. It also affected the fingers bilaterally to the extent that he stopped writing and started using keyboard to type instead.

Treatment induced neuropathy of diabetes (TIND) or insulin neuritis was suspected. He was regularly reviewed by a neurologist. Electrophysiological investigations confirmed evidence of an axonal length-dependent sensorimotor polyneuropathy with some axonal loss and demyelination. Various symptom modifying drugs, including amitriptyline, duloxetine, pregabalin and gabapentin were tried. A combination of duloxetine and gabapentin gradually alleviate the symptoms over a period of 6 months. Duloxetine dose was gradually titrated to 600 mg BD, and gabapentin to 1200mg, 900 mg, 1200 mg. He continued to have hypersensitive feet, which he described as “even soft touch can induce a sense of discomfort, but stronger pressure on the feet provokes actual pain”. After 18 months, these sensory complaints are now only limited to the soles of his feet. His balance is slightly impaired but he starts playing football again. He experiences pain in his fingers only intermittently when blood glucose is above 15 mmol/l.

The improvement in symptoms was reflected in the repeat electrophysiological parameters carried out 6 months later. The conduction velocities of the sensory and motor nerves improved and there were no signs of active and ongoing axonal loss. Electrophysiological studies at 18 months showed further improvement with a mild length-dependent predominantly sensory (more than motor) peripheral neuropathy. His diabetes control is well maintained with HbA1c of 55 mmol/mol (7.2%).

Discussion

TIND is an iatrogenic painful sensory neuropathy affecting small peripheral nerves1. It is characterised by acute neuropathic pain and autonomic dysfunction within 8 weeks of initiation of treatment. Previously, it was commonly known as insulin neuritis. The term TIND was proposed in 2010 after recognition that neuritis was not only secondary to insulin treatment but also due to a rapid reduction in glycaemic control2. It is usually seen in people with sudden decrease in HbA1c of more than 2% over 3 months2-4.

In literature, TIND is reported in both type 1 and type 2 diabetes regardless of age, duration of diabetes or type of treatment; insulin, oral hypoglycaemic agents or severe dietary restrictions4. Review of the case reports demonstrated that TIND tends to affect younger men and people with T1DM within 5 years of diagnosis5. The rate of change in glycaemic control appears to be directly proportional to the absolute risk of developing TIND6.

Pathogenesis of this condition is not fully understood. The proposed mechanisms includes endoneurial ischaemia, hypoglycaemic microvascular neuronal damage and regenerating nerve firing. Although the primary clinical manifestation is neuropathic pain, concurrent development of autonomic dysfunction, and worsening of other microvascular complications can occur7-9. Neuropathic pain affecting limbs can be very disabling and autonomic neuropathy resulting in arrhythmia can be life threatening in this condition. Neuropathic pain is characteristically burning in nature, predominantly affecting lower limbs. It can also affect other areas including the trunk, abdomen, or can be more generalised with allodynia.

The prevalence is unknown, but is seen in up to 10% of the cases referred for evaluation of diabetic neuropathy in tertiary centres3. It is more common than reported and hence clinicians should have a low threshold of suspicion in diagnosing this condition. Although optimising diabetes control is the priority in managing diabetes, mainly to reduce the risk of diabetes related complications, all health care professionals should be mindful that aggressive treatment with sudden reduction in glycated haemoglobin can result in more treatment related adverse outcomes as evident in this case and hence is not desirable.

At present, there are no consensus guidance on management of TIND. The mainstay of treatment is symptom management with analgesia. A combination of medications including anticonvulsant (pregabalin, gabapentin), anti-depressants (amitriptyline, duloxetine) or other simple analgesia have been used in case studies5. Initially, pain can be resistant to treatment despite use of multiple analgesia. Remission is observed in 60% of patients in case studies5. Other than symptom control with analgesia, the only known treatment for TIND to date is maintaining stable glucose level. Therefore, the focus of patient care should shift to prevention of development of TIND by avoiding rapid intensification of diabetes treatment or rapid correction of hyperglyacemia10.

Conclusion

TIND should be considered in any patients who develop neuropathy acutely following rapid correction of hyperglycaemia regardless of age, type or duration of diabetes or type of diabetes treatment. At present, optimising analgesia and stabilising diabetes control are the only supportive treatments available but most cases are reversible. Underlying mechanism of TIND is poorly understood and therefore further research is needed to elucidate better understanding of pathophysiology to prevent development of TIND and to guide treatment recommendations.

This case highlights that rapid intensification of glycaemic control with sudden improvement in glycated haemoglobin can result in undesirable outcome. As TIND can be a very disabling condition affecting quality of life, both patients and clinicians should have heightened awareness of consequences of rapid intensification of diabetes control.

References

1. Chandler, E., et al, (2019) Treatment induced neuropathy of diabetes (TIND) in pediatrics: a case report and review of the literature. J Clin Endocrinol Metab. 2019; 105:395–8.
2. Hwang Y.T., Davies, G., (2016) ‘Insulin neuritis’ to ‘treatment-induced neuropathy of diabetes’: new name, same mystery. Pract Neurol. 2016; 16:53–5.
3. Gibbons, C.H., (2017) Treatment-induced neuropathy of diabetes. Curr Diab Rep. 2017; 17: 127.
4. Gibbons, C.H., Freeman, R., (2010) Treatment-induced diabetic neuropathy: a reversible painful autonomic neuropathy. Ann Neurol. 2010; 67: 534-41.
5. Quiroz-Aldave, J.E., et al, (2023) Treatment-induced neuropathy of diabetes; underdiagnosed entity. Lancet Neurol. 2023; 22:201-2.
6. Gibbons, C. H., Freeman, R., (2015) Treatment-induced neuropathy of diabetes: an acute, iatrogenic complication of diabetes. Brain. 2015; 138: 43–52.
7. Asbury, A. K., Fields H.L., (1984) Pain due to peripheral nerve damage: a hypothesis. Neurology 1984; 34: 1587–90.
8. Tesfaye, S., et al., (1996) Arterio-venous shunting and proliferating new vessels in acute painful neuropathy of rapid glycaemic control (insulin neuritis). Diabetologia 1996; 39: 329–35.
9. Bain, S.C., et al., (2019) Worsening of diabetic retinopathy with rapid improvement in systemic glucose control: a review. Diabetes Obes Metab 2019; 21: 454–66.
10. Elafros, M.A., et al., (2023) Treatment-induced neuropathy of diabetes; underdiagnosed entity. Lancet Neurol. 2023; 22:201-2.

Authors

Hnin Thandar Aung1, Shar Younn Latt 2, Zin Zin Htike 3

1. Senior House Officer, MBBS (Yangon), Nottingham Universtiy Hosptials, Nottingham, UK
2. Senior House Officer, MBBS (Yangon), MSc (Cardiology & Stroke, University of Herfordshire), Nottingham Universtiy Hosptials, Nottingham, UK
3. Consultant Physician, MBBS (Yangon), MRCP (Endocrinology), FRCP, PhD (University of Leicester), Training Programme Director (Endocrinology, NHSE Eastmidland Deanery) Nottingham Universtiy Hosptials, Nottingham, UK

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button