Persistent hypothermia after intrathecal morphine: case report and literature review

  • Kenneth F. Ryan
  • James W. Price
  • C. Brian Warriner
  • Peter T. ChoiEmail author
Case Reports/Case Series



To describe a case of persistent hypothermia following spinal anesthesia with intrathecal morphine.

Clinical features

Following elective right total knee arthroplasty under spinal anesthesia with isobaric 0.5% bupivacaine 11 mg, fentanyl 15 μg, and preservative-free morphine 150 μg, a 57-yr-old female (93.5 kg, 151 cm) developed postoperative hypothermia with a nadir rectal temperature of 33.6°C four hours after surgery. At times, her temperature could not be measured by tympanic, temporal arterial, oral, axillary, or rectal routes. In spite of the low temperature, the patient complained of feeling hot and was diaphoretic without shivering. With the exception of her temperature, her vital signs were normal postoperatively, and aside from hyperglycemia, complete blood count, electrolytes, thyroid-stimulating hormone, serum cortisol, troponin, and twelve-lead electrocardiogram were normal. Her temperature did not respond to warming efforts with a forced-air warming blanket, infusion of warmed intravenous crystalloid, and hourly bladder irrigation with warm saline through an indwelling urinary catheter. Normothermia returned after she received a small dose of sublingual lorazepam eight hours after surgery. The remainder of her postoperative stay was uneventful.


Patients undergoing spinal anesthesia with intrathecal morphine may develop postoperative hypothermia that is resistant to warming measures. This complication may be treated successfully with lorazepam.


Morphine Total Knee Arthroplasty Bupivacaine Spinal Anesthesia Intrathecal Morphine 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Hypothermie persistante après administration intrathécale de morphine: étude de cas et revue de la littérature



Décrire un cas d’hypothermie persistante après anesthésie médullaire avec administration intrathécale de morphine.

Caractéristiques cliniques

Après arthroplastie totale programmée du genou droit sous anesthésie médullaire utilisant de la bupivacaïne 0,5 % isobare 11 mg, du fentanyl 15 μg et de la morphine 150 μg sans agent de conservation, une femme âgée de 57 ans (93,5 kg, 151 cm) a présenté une hypothermie postopératoire avec une température rectale descendant jusqu’à 33,6 °C quatre heures après l’opération. Par moment, sa température n’a pas pu être mesurée par voie tympanique, orale, axillaire, ou rectale, ou par le biais de l’artère temporale. En dépit de cette température basse, la patiente s’est plainte d’une sensation de chaleur et présentait une diaphorèse sans frissonnement. À l’exception de sa température, ses signes vitaux postopératoires étaient normaux et hormis une hyperglycémie, la formule sanguine complète, les électrolytes, l’hormone thyréotrope, le cortisol sérique, la troponine et un électrocardiogramme à 12 dérivations étaient normaux. Sa température n’a pas réagi aux efforts de réchauffement avec une couverture chauffante à ventilation forcée, une perfusion intraveineuse de cristalloïdes réchauffés et une irrigation vésicale de sérum physiologique chaud réalisée chaque heure au moyen d’une sonde urinaire à demeure. La patiente est redevenue normothermique huit heures après l’intervention après avoir reçu une petite dose sublinguale de lorazépam. Le reste de son hospitalisation postopératoire n’a été marquée par aucun événement particulier.


Les patients subissant une anesthésie médullaire avec administration intrathécale de morphine peuvent développer une hypothermie postopératoire résistante aux mesures de réchauffement. Cette complication peut être traitée avec succès par du lorazépam.

Intrathecal morphine is an analgesic commonly used in spinal anesthesia for a variety of surgical procedures. The advantages of intrathecal morphine include its opioid sparing effect in the postoperative period,1 targeted site of action in the central nervous system, and extended duration of action mediated by its effect on opioid receptors in the spinal cord.2 Intrathecal doses of morphine can also be associated with pruritis, urinary retention, nausea, and delayed respiratory depression.2,3 While hypothermia can be a common consequence of spinal anesthesia due to vasodilatation and the loss of the shivering reflex,4 several anecdotal reports describe persistent hypothermia in the obstetrical population after spinal anesthesia using intrathecal morphine.5-9 This case report describes persistent hypothermia following spinal anesthesia using intrathecal morphine for total knee arthroplasty and the return to normothermia after sublingual lorazepam.

Case presentation

We obtained the patient’s written consent to publish this case report. A 57-yr-old Caucasian woman (93.5 kg, 151 cm) with a ten-year history of bilateral knee osteoarthritis underwent an elective right total knee arthroplasty at the University of British Columbia Hospital. The patient’s medical history included a long history of smoking, stable mild chronic airflow limitation without recent exacerbation, occasional gastroesophageal reflux, dyslipidemia, and morbid obesity (body mass index 41.5 kg·m−2). Her routine medications were oral diclofenac and rosuvastatin and inhaled fluticasone-salmeterol. Aside from postoperative nausea and vomiting, she had not experienced any adverse effects from her previous anesthetics (two general anesthetics and two spinal anesthetics). Details from her prior anesthetics were unavailable.

On the morning of her surgery, the patient’s vital signs were normal (oxygen saturation [SpO2] 98%, respiratory rate [RR] 16 breaths·min−1, blood pressure [BP] 130/70 mmHg, oral temperature 36.6°C) except for a sinus bradycardia (heart rate [HR] 50 beats·min−1). Prior to anesthesia, she received cefazolin 2 g iv, oral acetaminophen 1,625 mg, oral ranitidine 150 mg, and inhaled nebulized salbutamol 2.5 mg. Upon arrival in the operating room, she received dexamethasone 4 mg iv and midazolam 1.5 mg iv. After application of standard anesthetic monitors and oxygen 8 L·min−1 via a face mask, we achieved spinal anesthesia with isobaric 0.5% bupivacaine 11 mg, fentanyl 15 μg, and preservative-free morphine 150 μg injected via a Quincke 22-G needle at the interspace between the third and fourth lumbar vertebrae. The patient was sedated with an intravenous infusion of propofol (average rate 100 μg·kg−1·min−1), and she received an intravenous infusion of room-temperature 0.9% saline 1,000 mL intraoperatively. Her routine vital signs were monitored with the exception of body temperature. Due to the speed of the surgeon, we used warmed blankets instead of a forced-air warming blanket to keep the patient warm. A tourniquet was applied above the operative site, inflated prior to incision, and deflated after skin closure. Intraoperative blood loss was negligible. The surgery was uneventful and the patient was transferred to the postanesthetic care unit (PACU) in stable condition. The total tourniquet, surgical, and intraoperative times were 45, 67, and 113 min, respectively.

On arrival in the PACU, her initial postoperative vital signs were SpO2 99%, RR 14 breaths·min−1, HR 70 beats·min−1, BP 100/50 mmHg, and temporal arterial temperature 35.9°C. Sensation to temperature was blocked up to the T2 dermatome, and she was nauseous and diaphoretic. Her nausea resolved after one dose each of dimenhydrinate 25 mg iv and prochlorperazine 5 mg iv. She remained diaphoretic during her PACU stay. The PACU nurses could not measure her temporal arterial, oral, or axillary temperatures after the initial measurement, but the patient stated that she felt hot. She was transferred to the ward 2.5 hr after surgery when she met our PACU discharge criteria (with the exception of temperature).

On the ward, the patient’s RR, HR, and BP remained normal. She was somnolent but was easily roused. Tympanic, temporal arterial, oral, and axillary temperatures could not be measured; rectal temperature was 34.1°C. Measurement of temperature remained difficult over the next five hours, and when measurable, rectal temperature fluctuated in the mid-33 to mid-34°C range with a nadir of 33.6°C four hours after surgery. In spite of the low temperature measurements, the patient remained diaphoretic without shivering and described feeling hot. Respiratory rate, HR, and BP remained normal. Aside from hyperglycemia (9.9 mmol·L−1), all other investigations (complete blood count, electrolytes, thyroid-stimulating hormone, serum cortisol, troponin, and twelve-lead electrocardiogram) were normal. We attempted to actively re-warm the patient using a forced-air warming blanket, an infusion of warmed intravenous crystalloid 1 L (via a fluid warmer) over an hour, and hourly bladder irrigation with warm saline through an indwelling urinary catheter.

In spite of our attempts to re-warm the patient, we were eventually unable to measure her temperature using any route (tympanic, temporal arterial, oral, axillary, and rectal) eight hours after surgery. We could not identify a possible etiology for the hypothermia except intrathecal morphine. We avoided using naloxone to reverse the effects of the intrathecal morphine as we were concerned about antagonizing its analgesia. Using PubMed, we identified a case series describing hypothermia in parturients after Cesarean delivery under spinal anesthesia with intrathecal morphine.5 The authors suggested an association between lorazepam and the resolution of hypothermia.5 Based on this report, we administered sublingual lorazepam 0.5 mg to the patient. Within 25 min, the patient was no longer diaphoretic or feeling warm. Her temporal arterial temperature was 35.6°C, and it slowly increased to normal levels by the next morning. The remainder of her postoperative course was uneventful.


We describe a case of mild persistent postoperative hypothermia after spinal anesthesia using intrathecal morphine 150 μg in a patient undergoing total knee arthroplasty. Normal body temperature was measured after treatment with a small dose of sublingual lorazepam. While spinal anesthesia causes a drop in body temperature due to vasodilatation and a loss of the shivering reflex,4 the resulting hypothermia is usually treated successfully with external re-warming techniques, such as increasing the room temperature, provision of dry linens and gowns, and the use of forced-air warming blankets. In this case, other causes of perioperative hypothermia were considered as causes for this patient’s hypothermia, including low ambient room temperatures, cold intraoperative intravenous fluid administration, high spinal blockade, and a relatively small body-to-surface area ratio in this short-statured patient.

We included measurement error in our early differential diagnosis of low temperature. In our PACU, the temporal artery is the initial site of temperature measurement. Infrared temporal artery thermometry is less accurate than invasive methods,10 but it is better than tympanic thermometry and is easy to use in the PACU. Diaphoresis may have impeded its ability to record a temperature in our patient. Measurements of low temperature using other thermometers and routes enabled us to rule out measurement error. We are uncertain why we eventually could not measure a temperature in our patient since our thermometers had measurement limits lower than 30°C (16°C, TemporalScanner™ TAT-5000, Exergen, Boston, MA, USA; 26.7°C, Sure Temp® Plus, Welch Allyn, Skaneateles Fall, NY, USA).1

After the resolution of the spinal sensory blockade, the maladaptive response of diaphoresis, the lack of shivering, and the failure of aggressive warming techniques suggested another cause. It is our view that intrathecal morphine was the cause of the persistent hypothermia and abnormal thermoregulatory response in our patient.

Persistent hypothermia after spinal anesthesia with intrathecal morphine has been described in six other case reports of 20 patients (Table),5-9,11 and another case has been described in a non-obstetrical patient.11 All of the patients reported nausea and diaphoresis; a number of them complained of feeling hot, and none of them shivered or felt cold. The nadir of the temperature measurements ranged from 33.2°C (tympanic)9 to 34.9°C (mean oral).8 The duration of hypothermia varied widely (2-22 hr).

Previous case reports of persistent hypothermia after spinal anesthesia using intrathecal morphine

Authors (Year)

Surgery (number of patients)

Spinal Anesthetic

Description of Symptoms

Lowest Temperature (Route)

Duration of Hypothermia

Kosai et al. (1992)11

Split thickness skin grafting (n = 1)

0.5% tetracaine 14 mg,

Nausea, sweating, feeling hot, no shivering

34.0°C (bladder)

>19 hr

0.125% phenylephrine 3.5 mg,

morphine 0.14 mg

Wishaw (1997)5

Cesarean delivery (n = 1)

0.5% bupivacaine 11 mg,

Flushed, feeling hot

33.6°C (aural)

2 hr*

fentanyl 10 μg,

morphine 0.25 mg

Peillon et al. (2002)6

Cesarean delivery (n = 2)

0.5% bupivacaine 8 - 10 mg,

No shivering, not cold

33.6 & 34.6°C (rectal)

5 hr & 4 hr

sufentanil 2.5 μg,

morphine 0.1 mg

Sayyid et al. (2003)7

Cesarean delivery (n = 1)

0.75% bupivacaine 12 mg,

Nausea, sweating

33.6°C (not reported)

2 hr*

morphine 0.2 mg

Hess et al. (2005)8

Cesarean delivery (n = 14)

bupivacaine 11.25 mg,

Nausea, sweating, feeling hot

34.9°C† (oral)

4 - 6 hr‡

fentanyl 25 μg,

morphine 0.25 mg

Fischer et al. (2006)9

Cesarean delivery (n = 1)

0.5% bupivacaine 10 mg,

Nausea, sweating, confusion, no shivering, not cold

33.2°C (tympanic)

22 hr

morphine 0.1 mg

* This patient became normothermic after receiving naloxone

Mean of the lowest temperatures for all 14 patients

This range of duration is reported for the four patients who did not receive lorazepam

In a randomized controlled trial of 60 parturients undergoing Cesarean delivery, Hui et al. compared the temperatures of those undergoing spinal anesthesia (0.5% bupivacaine 10-12 mg) with intrathecal morphine 150 μg with those undergoing spinal anesthesia (0.5% bupivacaine 10-12 mg) without morphine (and an equivalent volume of intrathecal 0.9% saline).12 Decreases in tympanic temperature occurred in both groups, but the morphine group experienced a larger decrease than the placebo group, i.e., 1.11 (0.61)°C vs 0.76 (0.39)°C, respectively; P = 0.01.12 The lowest tympanic temperatures were 34.3°C and 35.2°C in the morphine and placebo groups, respectively.12

The effect of morphine on thermoregulation is species-dependent, with morphine causing hyperthermia exclusively in some animals (e.g., rats)13 and hypothermia in others (e.g., dogs).14 Its effect on thermoregulation is also dose-dependent, with smaller doses causing hyperthermia and larger doses causing hypothermia.15 Results from these animal studies suggest that intrathecal morphine-induced hypothermia may be due to a central effect of morphine on the thermoregulatory centres in the hypothalamus. Morphine may play a role in hypothermia by altering the temperature set point in the hypothalamus. This may result in maladaptive changes, such as sweating and an absence of shivering that may exacerbate the hypothermia and contribute to the subjective sense of feeling warm, which is opposite to the effect of chills associated with a febrile patient.16-18 The mechanism for morphine’s effect on the hypothalamus has yet to be clearly defined.

The roles of various receptors in thermoregulation are still being defined. In studies comparing the effect of morphine vs meperidine (an opioid used to treat postoperative shivering) on abolishing shivering, a relationship is suggested between the loss of shivering and the intensity of kappa receptor agonism, with the correlation being greater with meperidine.19 However, tramadol has also been used to treat shivering, and its opioid effect is at the mu receptor, similar to morphine.20

The dorsomedial hypothalamus has been identified as a location of thermoregulation in the brain. In rats, fentanyl acts through the dorsomedial hypothalamus and causes initiation of non-shivering thermogenesis by metabolism of brown fat. This is thought to be a stress response in rats. There are substantially more kappa receptors than mu receptors in the dorsomedial hypothalamus,21 and this difference may account for the greater efficacy of meperidine compared with morphine to abolish shivering. Perhaps an intrathecal dose of morphine could saturate receptors locally in the dorsomedial hypothalamus with a similar effect to meperidine. Morphine may also alter the temperature set point in the hypothalamus causing the body to feel hot and then sweat to adapt to the heat. A mechanism for this has not been described. In the anterior hypothalamus, morphine inhibits “cool-sensitive interneurons” via serotonin-secreting neurons and causes a decrease in metabolic heat production.22 Understanding the interplay of morphine’s effect on mu receptors in different parts of the hypothalamus may be the key to define the mechanism by which it causes hypothermia in humans.

Hess et al. described a return (within 90 min) to normothermia in eight of the ten parturients who received lorazepam.8 We also observed a rapid increase in our patient’s body temperature. Lorazepam binds to benzodiazepine receptors, which are subunits of gamma-aminobutyric acid-A (GABAA) receptors, and enhances the effect of GABA on GABAA receptors. Although benzodiazepine receptors are found mostly in the cerebral cortex, they are also found in the hypothalamus. Binding of benzodiazepine receptors and its effect on GABA-related activity in the hypothalamus may modulate thermoregulation. The role of GABA receptors in thermoregulation is supported by animal studies in which the administration of GABAergic substances in the central nervous system (CNS) affects thermoregulation.23,24 In humans, it remains unclear whether GABA activity results in hyperthermia or hypothermia. Other investigators have reported hypothermia in patients receiving valproic acid, which is a GABA receptor agonist.25 The role of GABAergic substances in thermoregulation requires further elucidation.

In summary, we observed persistent hypothermia in a patient after spinal anesthesia using intrathecal morphine. Normothermia returned after administration of lorazepam. The mechanisms by which morphine and lorazepam affect thermoregulation remain unclear.




We thank the patient for granting permission to publish this case report. We also thank Dr. Steven Sutcliffe and the nurses from the University of British Columbia Hospital PAR and the University of British Columbia Hospital Surgical Observation Unit for their care of the patient.

Funding source


Conflicts of interest

None declared.


  1. 1.
    Graham D, Russell IF. A double-blind assessment of the analgesic sparing effect of intrathecal diamorphine (0.3 mg) with spinal anaesthesia for elective caesarean section. Int J Obstet Anesth 1997; 6: 224-30.PubMedCrossRefGoogle Scholar
  2. 2.
    Maylan N, Elia N, Lysakowski C, Tramer MR. Benefit and risk of intrathecal morphine without local anaesthetic in patients undergoing major surgery: meta-analysis of randomized trials. Br J Anaesth 2009; 102: 156-67.CrossRefGoogle Scholar
  3. 3.
    Rathmell JP, Lair TR, Nauman B. The role of intrathecal drugs in the treatment of acute pain. Anesth Analg 2005; 101: S30-43.PubMedCrossRefGoogle Scholar
  4. 4.
    Kurz A, Sessler DI, Schroeder M, Kurz M. Thermoregulatory response thresholds during spinal anesthesia. Anesth Analg 1993; 77: 721-6.PubMedGoogle Scholar
  5. 5.
    Wishaw K. Hypothermia associated with subarachnoid morphine. Anaesth Intensive Care 1997; 25: 586.PubMedGoogle Scholar
  6. 6.
    Peillon P, Dounas M, Lebonhomme JJ, Guittard Y. Severe hypothermia associated with cesarean section under spinal anesthesia (French). Ann Fr Anesth Reanim 2002; 21: 299-302.PubMedGoogle Scholar
  7. 7.
    Sayyid SS, Jabbour DG, Baraka AS. Hypothermia and excessive sweating following intrathecal morphine in a parturient undergoing cesarean delivery. Reg Anesth Pain Med 2003; 28: 140-3.PubMedGoogle Scholar
  8. 8.
    Hess PE, Snowman CE, Wang J. Hypothermia after cesarean delivery and its reversal with lorazepam. Int J Obstet Anesth 2005; 14: 279-83.PubMedCrossRefGoogle Scholar
  9. 9.
    Fischer MO, Dequire PM, Kalem A, Gerard JL, Plaud B. Hypothermia after spinal anaesthesia: implication of morphine? (French). Ann Fr Anesth Reanim 2006; 25: 296-8.PubMedCrossRefGoogle Scholar
  10. 10.
    Kimberger O, Cohen D, Illievich U, Lenhardt R. Temporal artery versus bladder thermometry during perioperative and intensive care unit monitoring. Anesth Analg 2007; 105: 1042-7.PubMedCrossRefGoogle Scholar
  11. 11.
    Kosai K, Takasaki M, Kawasaki H, Nagata N. Hypothermia associated with intrathecal morphine. J Anesth 1992; 6: 349-52.PubMedCrossRefGoogle Scholar
  12. 12.
    Hui CK, Huang CH, Lin CJ, Lau HP, Chan WH, Yeh HM. A randomised double-blind controlled study evaluating the hypothermic effect of 150 μg morphine during spinal anaesthesia for caesarean section. Anaesthesia 2006; 61: 29-31.PubMedCrossRefGoogle Scholar
  13. 13.
    Shen Z, Yanase M, Kanosue K, Nakayama T. Effect of periaqueductal morphine injection on thermal response in rats. Jpn J Physiol 1986; 36: 485-96.PubMedCrossRefGoogle Scholar
  14. 14.
    Sabbe MB, Grafe MR, Mjanger E, Tiseo PJ, Hill HF, Yaksh TL. Spinal delivery of sufentanil, alfentanil, and morphine in dogs: physiologic and toxicologic investigations. Anesthesiology 1994; 81: 899-920.PubMedCrossRefGoogle Scholar
  15. 15.
    Clark SM, Murphy MT, Lipton JM, Clark WG. Effects of morphine on body temperature of squirrel monkeys of various ages. Brain Res Bull 1983; 10: 305-8.PubMedCrossRefGoogle Scholar
  16. 16.
    Clark WG. Influence of opioids on central thermoregulatory mechanisms. Pharmacol Biochem Behav 1979; 10: 609-13.PubMedCrossRefGoogle Scholar
  17. 17.
    Su CF, Liu MY, Lin MT. Intraventricular morphine produces pain relief, hypothermia, hyperglycaemia and increased prolactin and growth hormone levels in patients with cancer pain. J Neurol 1987; 235: 105-8.PubMedCrossRefGoogle Scholar
  18. 18.
    Spencer RL, Hruby VJ, Burks TF. Alteration of thermoregulatory set point with opioid agonists. J Pharmacol Exp Ther 1990; 252: 696-705.PubMedGoogle Scholar
  19. 19.
    Tsai YC, Chu KS. A comparison of tramadol, amitriptyline, and meperidine for postepidural anesthetic shivering in parturients. Anesth Analg 2001; 93: 1288-92.PubMedCrossRefGoogle Scholar
  20. 20.
    Kurz M, Belani KG, Sessler DI, et al. Naloxone, meperidine, and shivering. Anesthesiology 1993; 79: 1193-201.PubMedCrossRefGoogle Scholar
  21. 21.
    Mansour A, Khachaturian H, Lewis ME, Akil H, Watson SJ. Anatomy of CNS opioid receptors. Trends Neurosci 1988; 11: 308-14.PubMedCrossRefGoogle Scholar
  22. 22.
    Oka T. 5-HT and narcotic-induced hypothermia. Gen Pharmacol 1978; 9: 151-4.PubMedCrossRefGoogle Scholar
  23. 23.
    Paredes RG, Agmo A. GABA and behavior: the role of receptor subtypes. Neurosci Biobehav Rev 1992; 16: 145-70.PubMedCrossRefGoogle Scholar
  24. 24.
    Morrison SF. 2010 Carl Ludwig Distinguished Lectureship of the APS Neural Control and Autonomic Regulation Section: Central neural pathways for thermoregulatory cold defense. J Appl Physiol 2011; 110: 1137-49.PubMedCrossRefGoogle Scholar
  25. 25.
    Zachariah SB, Zachariah A, Ananda R, Stewart JT. Hypothermia and thermoregulatory derangements induced by valproic acid. Neurology 2000; 55: 150-1.PubMedGoogle Scholar

Copyright information

© Canadian Anesthesiologists' Society 2011

Authors and Affiliations

  • Kenneth F. Ryan
    • 1
    • 2
  • James W. Price
    • 1
    • 2
  • C. Brian Warriner
    • 1
    • 2
  • Peter T. Choi
    • 1
    • 2
    Email author
  1. 1.Department of AnesthesiaVancouver AcuteVancouverCanada
  2. 2.Department of Anesthesiology, Pharmacology and TherapeuticsUniversity of British ColumbiaVancouverCanada

Personalised recommendations