The ofcial journal of the Australian Dental Association

Australian Dental Journal

SCIENTIFIC ARTICLE
Australian Dental Journal 2013; 58: 8993 doi: 10.1111/adj.12029

The healing of dental extraction sockets in patients with Type 2 diabetes on oral hypoglycaemics: a prospective cohort
S Huang,* H Dang,* W Huynh,* PJ Sambrook,* AN Goss*
*Oral and Maxillofacial Unit, Royal Adelaide Hospital, South Australia. School of Dentistry, The University of Adelaide, South Australia.

ABSTRACT
Background: The aim of this study was to determine whether there is a difference in delayed healing following dental extractions for Type 2 diabetics on oral hypoglycaemics and non-diabetic patients. Methods: Prospective patients referred for dental extractions were recruited into two groups: known diabetics and nondiabetics with no conditions associated with poor healing. All had a random blood glucose level (BGL). Extractions were performed using local anaesthesia. Delayed healing cases were identied and statistical evaluation performed to identify risk factors. Results: There were 224 Type 2 diabetics on oral hypoglycaemics (BGL 7.51, range 4.117.4) and 232 non-diabetics. The diabetic group were older, more males and less smokers than the control group. Twenty-eight patients, 12 (5%) diabetic and 16 (7%) control group, had socket healing delayed for more than one week but all healed in four weeks. There were no statistical differences between delayed healing and age, gender, diabetic state, BGL or smoking. The younger control group had more healing problems. Conclusions: The traditional view that diabetics have increased delayed healing was not supported. Type 2 diabetics on oral hypoglycaemics should be treated the same as non-diabetic patients for extractions.
Keywords: Diabetes, Type 2, blood glucose levels, extractions, delayed healing, prospective cohort controlled study. Abbreviations and acronyms: BGL = blood glucose level; IFG = impaired fasting glucose; IGT = impaired glucose tolerance; OMS = oral and maxillofacial surgery. (Accepted for publication 16 July 2012.)

INTRODUCTION Traditionally in dentistry diabetics are considered to have increased healing problems related to dental extractions, periodontal surgery and wearing ill-tting dentures.1 They are also considered more likely to have infections. Although this may be so for poorly controlled Type 1 diabetics, there is only anecdotal support for this view for Type 2 diabetics on oral hypoglycaemics. There are no evidence based studies such as case controlled cohort studies for dental surgery in Type 2 diabetes. This is an important evidence based deciency as Type 2 diabetics constitute 90% of all diabetic patients. Recently after the completion of the data collection of the present study, a report comparing well controlled with poorly controlled diabetics, as measured by blood glucose level (BGL), haemoglobin
2013 Australian Dental Association

Alc (HbAlc) and endorganic scores was published. It found no difference in healing.2 Diabetes is a common metabolic disorder characterized by an inability to regulate blood glucose due to insulin deciency or resistance. Type 1 diabetes (previously known as insulin-dependent, juvenile or childhood-onset) is characterized by decient insulin production whereas Type 2 diabetes (formerly called non-insulin-dependent or adult-onset) results from relative insulin deciency and tissue insulin resistance causing abnormal BGLs despite secondary hyperinsulinaemia.3,4 In the AusDiab Study of 2002 it was found that 7.4% of Australian adults were diabetic and a further 16.4% were prediabetic.5 For every two known diabetics it has been found that there is at least one unknown diabetic. This number is increased by the two prediabetic states of impaired fasting glucose (IFG) and impaired glucose tolerance (IGT).
89

S Huang et al. Patients with prediabetes do not meet the criteria for being diagnosed with diabetes but have glucose levels higher than those considered normal.6 For IFG, it is a fasting BGL of 6.17 mmol/L and for IGT, it is a non-fasting BGL of 7.811.0 mmol/L. Each year, 310% of people with prediabetes will go on to develop diabetes.4 The clinically relevant BGLs are 7.8 mmol/L for prediabetes and 11.0 mmol/L for diabetes.7 Another useful measure of long-term glycaemic control is the haemoglobin Alc (HbAlc) test where the target is 6.57.0%, with above 8.0% being an indicator of poorer control. It is recommended that dentists determine the stability of known diabetics by means of the BGL and HbAlc, either by patient history, advice from their medical practitioner or by directly performing the tests prior to commencing surgery.7 Poor glycaemic control predisposes to development of a range of complications that have been broadly categorized as macrovascular, microvascular and neuropathic.8 Microcirculatory deciencies, in particular, can have signicant bearing on wound healing following surgical procedures. An intact microcirculation is required for tissue nutrition, removal of waste products, inammatory responses and temperature regulation.9 In diabetics, changes to the capillaries such as thickening of the basement membrane result in altered permeability, impeded migration of leucocytes and impaired hyperaemia, causing underperfusion during tissue stress and tissue hypoxia.9,10 These changes can adversely affect the outcome of surgery, resulting in poor wound healing and wound infection.11,12 This has been most clearly documented in relation to cardiothoracic procedures where poorly controlled diabetic patients undergoing coronary bypass surgeries experienced increased morbidity and mortality,13 increased rates of postsurgical infections14 and worse hospital outcomes.15,16 Conversely, maintenance of tight perioperative glycaemic control for coronary bypass patients has been shown to signicantly decrease infections and other adverse outcomes.17 It has been shown that diabetic complications can occur during prediabetes, particularly microangiopathy such as diabetic retinopathy.18 It is for these reasons that known diabetics are offered counselling on nutrition and lifestyle including smoking cessation.5 In dentoalveolar surgery, diabetic patients could be expected to suffer similar complications to those observed in other surgical procedures. However, the oral environment with the forces of mastication, high bone turnover, high vascularity, saliva and the constant reservoir of microorganisms is distinct from other parts of the body, thereby making generalizations from other surgical sites limited.19 There have been a limited number of studies using experimentally induced diabetes in rats. These animals have an uncontrolled insulin dependent diabetic state
90

and not surprisingly dental extraction wounds heal poorly, often with alveolar destruction.20 However, this does not represent a current clinical situation in an advanced country such as Australia unless both the patient and their treating dentist completely mismanage an unstable Type 1 diabetic state. This may occasionally occur and the patient ends up in hospital, requiring specialist management. Such cases are not reported but the two consultant oral and maxillofacial surgeons (ANG, PJS) in this study have encountered such cases. The aim of this study is to investigate and compare the difference in healing between Type 2 diabetics and non-diabetics, and whether differences in a random BGL are of any signicance. METHODS This prospective study was performed in the exodontia clinic of the oral and maxillofacial surgery (OMS) unit in the outpatient clinic at the Adelaide Dental Hospital, South Australia. Appropriate ethics approval from the Royal Adelaide Hospital was obtained (RAH REC 091115). Patients 18 years and over who were referred for extractions under local anaesthesia during the academic year 2010 and had a detailed medical history were approached to give written consent to be involved in the study. Exclusion criteria related to known conditions which may impair healing of extraction namely: Type 1 diabetes and insulin dependent Type 2 diabetes; HIV/AIDS; chemotherapy; systemic steroids; irradiation to the head and neck; dental infections with systemic involvement; bisphosphonates; anticoagulant and antiplatelet treatment; and major benign or malignant pathology within the jaws. Patients unable to give consent through physical or mental disability were also excluded. All patients had a random BGL taken after administration of the local anaesthetic (2% lignocaine with 1 in 80 000 adrenaline) using a glucometer (Optimum Xceed Abbott Diabetes Care, Doncaster, VIC, Australia). Patients were initially assigned to two groups: known Type 2 non-insulin dependent diabetics on oral hypoglycaemics and the control group without conditions known to impair healing. The intra-alveolar extraction of erupted teeth was then performed with forceps and elevators by nal year dental students under the supervision of OMS staff. Antibiotics were only prescribed if there was clear evidence of localized acute odontogenic infection with pus present. Patients with spreading infection were excluded. Patients were offered follow-up review at one week or if they declined, advised to contact or return to the OMS clinic if they had problems.
2013 Australian Dental Association

Dental extractions for diabetics The random BGLs of the control group were checked. Those with a BGL above 7.8 mmol/L were counselled by OMS staff and referred in writing to their general medical clinic.4,7 Diabetic patients with a BGL above 11 mmol/L were also advised and counselled.7 The results of completed questionnaires were entered into a Microsoft spreadsheet on a standalone PC at the end of each day. The patient records were reviewed at four weeks by one researcher (SH) to determine who had returned or reported delayed healing. This included dry socket, bony sequestra or excess granulation tissue after one week. The patients medical practitioner was contacted at eight weeks to determine the outcome of investigations for those non-diabetic patients with a BGL above 7.8 mmol/L. Patients who demonstrated to be diabetic were reassigned to the diabetic group. Data were analysed using SAS software Version 9.3 (SAS Institute Inc, Cary, NC, USA). The odds of delayed healing were compared between the diabetics and non-diabetic groups using logistic regression, with results expressed as odds ratios with 95% condence intervals. Logistic regression was also used to test the association between perioperative BGL and the odds of delayed healing. Lastly, a multivariate logistic regression model was used to test whether age, gender and smoking had any effect on delayed healing. Throughout all analyses a two-tailed p-value of 0.05 was used to indicate statistical signicance. RESULTS Five hundred and ten patients who met the inclusion criteria for the study were approached to participate. Thirty (6.0%) declined to give consent, all were nondiabetics and received routine extractions outside of the trial. Of the remaining 480, 222 were known Type 2 diabetics on oral hypoglycaemics and not taking insulin, with an average BGL of 7.5 mmol/L, range 4.1 to 17.4. Two hundred and fty-eight were in the control group but of these, 32 had a random BGL of above 7.8 mmol/L. All of these were advised to seek medical advice and were given a written referral with a copy directly to the medical clinic. One was found to be a diabetic, one a prediabetic and Table 1. Demographics and overall outcome
Group n Mean Non-diabetics Diabetics 232 224 46.8 63.9 Age Range 1688 1987 M F M F Gender n 105 127 141 83 % 45 55 63 37 Mean 5.2 7.5 BGL Range 1.99.2 4.117.4 Abnormal BGL n 2 86 % 0.86 38.4 n 90 43 Smoker % 38.8 19.2 Delayed healing n 16 12 % 6.9 5.4

these were reassigned to the diabetic group, making a total of 224 in the diabetic group. Six were found not to be diabetic or prediabetic and remained in the control group, making a total of 232. Twenty-four patients, despite verbal and written advice, did not seek medical attention. Of these, six made appointments but did not attend. One had severe mental issues and was admitted to an appropriate facility and found not to be diabetic. Seventeen were not known to the medical clinic they nominated and had not been seen. These 24 patients were excluded from the trial. Twenty-eight patients had healing delayed beyond one week; 12 (5%) in the diabetic group and 16 (7%) in the non-diabetic control group. All had fully healed within four weeks. There were no cases of osteomyelitis or osteonecrosis of the jaws. The age, gender, BGL, smoking status and adverse outcomes are presented in Table 1. The diabetic group was on average 17 years older with more males and less smokers. They had less delayed healing events. The relationship of the delayed healing to BGL is presented in Table 2. The relationship of the delayed healing to smokers is presented in Table 3. Evaluation of the association between diabetic status and adverse outcomes showed that while the odds were 30.9% higher in the non-diabetic group, 16 (7%) vs 12 (5%), the difference between groups was not statistically signicant (p = 0.49). In terms of the association between BGL and adverse outcomes, for every unit increase in BGL the odds of adverse outcomes increased by 0.9%, though this was also not statistically signicant (p < 0.93). An assessment of other factors showed that there was no difference in the odds of adverse outcomes according to age (p = 0.78), gender (p = 0.98) and smoking (p = 0.24) status. DISCUSSION This study shows that there was no statistically signicant difference in post-extraction outcome between Type 2 diabetics on oral hypoglycaemics and the control group. The number of previously unrecognized diabetics found in this study was low. Only 2 (25%) of the 8 patients with elevated BGL, whose outcome

2013 Australian Dental Association

91

S Huang et al. Table 2. Relationship between delayed healing and BGL

BGL Below 7.8 7.811 Above 11 Diabetic 7 (5.1%) 5 (6.3%) 0 (0%) Non-diabetic 16 (7.0%) 0 (0%) 0 (0%)

Table 3. Relationship between delayed healing and smokers

Smokers Yes No Diabetic 1 (2.3%) 11 (6.1%) Non-diabetic 4 (4.4%) 12 (8.5%)

was known, were subsequently proven to have diabetes or prediabetes. The 24 with elevated BGL who did not attend medical follow-up had an unknown status. The AusDiab study in 2002 found an incidence of 3.7% unknown diabetics and 16.4% with prediabetes, or 20%, in an adult Australian population.5 The AusDiab study was very large and is thus difcult to directly compare to this relatively small sample. However, the numbers are of similar order. The study population is different to the usual general dental population in that all the patients were referred to the exodontia clinic of a specialist OMS unit. The referral base is from the community dental service, the public outpatient clinics of the Royal Adelaide Hospital and other major teaching hospitals. As such all the patients have disability or other government health care cards. Previous audits have shown that as well as difcult extractions, over two-thirds are signicantly medically compromised. However, the initial assessment and extractions are performed by nal year dental students with close supervision by OMS staff. Only 2 of the 480 patients, 1 from each group, required formal surgical extractions. In the non-diabetic control group, 3 had acute oro-antral stulae created from maxillary extractions and all were immediately closed by the OMS staff without sequelae. Besides their medical complexity, frequently including mental health issues, many were patients who sought only extractions for dental treatment. Most were not interested in attending follow-up appointments citing transport, parking, travelling and waiting time issues as reasons. Thus some patients who initially had difculty may have attended their local community clinic or general medical practitioners. However, if there were ongoing issues the patient would have been referred back. There are no other public OMS services available to them. These patient attitudes to their overall health were reected in the
92

24 of 32 with abnormally high BGLs who did not seek follow-up from their general medical practitioner despite being encouraged to do so. Therefore, this study group represents a more challenging population than those commonly seen in a private general practice by general dentist. The original plan for the study was that the diabetic and non-diabetic groups would be of similar age and gender, but this did not eventuate. The diabetic group was older on average by 17 years and there were more males. Increased age and male gender are risk factors for delayed healing.21 Conversely, more of the control group were smokers, also an important factor in delayed healing.21 Although there were no statistical differences, the clinical impression was that the diabetic group had less delayed healing than the younger, more female control group. Overall, the study population had 29.2% smokers, which is higher than the recently reported gure of 19% for the adult Australian population.22 This probably relates to the lower socio-economic status of the study population. However, there was an important difference between the non-diabetic group of whom 39% smoked and the diabetic group with only 19% of smokers. This reects the effort put in by diabetic educators as smoking and diabetes result in a high risk of cardiovascular and peripheral vascular disease.5 At the time of development and implementation of this study there were no prospective cohort studies on the effect of diabetes on extraction wound healing. Recently, a broadly similar study has been published which compared the relationship of glycaemic control between well and poorly controlled diabetics.2 They found no statistically signicant difference between the two groups. This is a similar outcome to the present study. However, on detailed analysis there are some methodological issues with the study by Aronovich et al.2 They combined both Type 1 and Type 2 diabetics and did not have a non-diabetic control group. Thus, the present study looks at a homogenous group of Type 2 diabetics on oral hypoglycaemics and compares them to a non-diabetic population treated by the same staff, in the same clinic over the same time period. This strengthens the specic value of the study to dental practice. It is concluded that there is similar healing between Type 2 diabetics on oral hypoglycaemics and nondiabetic patients. Special precautions including warnings about adverse healing and prophylactic antibiotics for routine extractions are not required. ACKNOWLEDGEMENTS We acknowledge the participation of the nal year BDS class of 2010, The University of Adelaide for this study. The support of the OMS clinic staff,
2013 Australian Dental Association

Dental extractions for diabetics particularly the registered nurses, is acknowledged. The statistical advice of Dr T Sullivan of the Data Management and Analysis Centre at the University of Adelaide is gratefully acknowledged. REFERENCES
1. Australian Research Centre for Population Oral Health. Special Topic No. 3 Diabetes and Oral Health. Available at: http:// www.arcpoh.adelaide.edu.au/dperu/special/diabetes/DiabetesA4.pdf. Accessed 20 March 2012. 2. Aronovich S, Skope LW, Kelly JP, Kyriakides TC. The relationship of glycemic control to the outcomes of dental extractions. J Oral Maxillofac Surg 2010;68:29552961. 3. World Health Organization. Diabetes Fact Sheet No. 312. Available at: http://www.who.int/mediacentre/factsheets/fs312/ en/. Accessed 15 March 2012. 4. Twigg SM, Kamp MC, Davis TM, Neylon EK, Flack JR. Prediabetes: a position statement from the Australian Diabetes Society and Australian Diabetes Educators Association. Med J Aust 2007;186:461465. 5. Dunstan DW, Zimmet PZ, Welborn TA, et al. The rising prevalence of diabetes and impaired glucose tolerance: the Australian Diabetes, Obesity and Lifestyle Study. Diabetes Care 2002;25: 829834. 6. Buysschaert M, Bergman M. Denition of prediabetes. Med Clin North Am 2011;95:289297. 7. Therapeutic Guidelines. Oral and Dental Version 1. Melbourne: Therapeutic Guidelines Limited, 2007:72. 8. Rosenberg CS. Wound healing in the patient with diabetes mellitus. Nurs Clin North Am 1990;25:247261. 9. Ekmektzoglou KA, Zografos GC. A concomitant review of the effect of diabetes mellitus and hypothyroidism in wound healing. World J Gastroenterol 2006;12:27212729. 10. Lioupis C. Effects of diabetes mellitus on wound healing: an update. J Wound Care 2005;14:8486. 11. Peleg AY, Weerarathna T, McCarthy JS, Davis TM. Common infections in diabetes: pathogenesis, management and relationship to glycaemic control. Diabetes Metab Res Rev 2007;23:313. 12. Jacober SJ, Sowers JR. An update on perioperative management of diabetes. Arch Intern Med 1999;159:24052411. 13. Outtara A, Lecombe P, Le Manach Y, et al. Poor intraoperative blood glucose control is associated with worsened hospital outcome after cardiac surgery in diabetic patients. Anesthesiology 2005;103:687694. 14. Latham R, Lancaster AD, Covington JF, Pirolo JS, Thomas CS. The association of diabetes and glucose control with surgicalsite infections among cardiothoracic surgery patients. Infect Control Hosp Epidemiol 2001;22:607612. 15. Halkos ME, Puskas JD, Lattouf OM, et al. Elevated preoperative hemoglobin A1c level is predictive of adverse events after coronary artery bypass surgery. J Thorac Cardiovasc Surg 2008;136:631640. 16. Halkos ME, Lattouf OM, Puskas JD, et al. Elevated preoperative hemoglobin A1c level is associated with reduced long-term survival after coronary artery bypass surgery. Ann Thorac Surg 2008;86:14311437. 17. Lazar HL, Chipkin SR, Fitzgerald CA, Bao Y, Cabral H, Apstein CS. Tight glycemic control in diabetic coronary artery bypass graft patients improves perioperative outcomes and decreases recurrent ischemic events. Circulation 2004;109:14971502. 18. Diabetes Prevention Program Research Group. The prevalence of retinopathy in impaired glucose tolerance and recent-onset diabetes in the Diabetes Prevention Program. Diabet Med 2007;24:137144. 19. Barasch A, Safford MM, Litaker MS, Gilbert GH. Risk factors for oral postoperative infection in patients with diabetes. Spec Care Dentist 2008;28:159166. 20. Devlin H, Garland H, Sloan P. Healing of tooth extraction sockets in experimental diabetes mellitus. J Oral Maxillofac Surg 1996;54:10871091. 21. Noroozi A-R, Philbert RF. Modern concepts in understanding and management of the dry socket syndrome: comprehensive review of the literature. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2009;107:3035. 22. Scollo MM, Winstanley MH, eds. Tobacco in Australia: facts and issues. 3rd edn. Melbourne: Cancer Council Victoria, 2008.

Address for correspondence: Professor Alastair N Goss School of Dentistry The University of Adelaide Adelaide SA 5005 Email: alastair.goss@adelaide.edu.au

2013 Australian Dental Association

93