Preoperative HbA1c and Postoperative Outcomes in Spine Surgery

A Systematic Review and Meta-Analysis

Xu Tao, MA; Abhijith V. Matur, MD; Paolo Palmisciano, MD; Fatu S. Conteh, MD; Louisa Onyewadume, MD; Henry O. Duah, MPH; Geet Shukla, BA; Phillip Vorster, BS; Sahil Gupta, BS; Sai S. Chilakapati, BS; Owoicho Adogwa, MD

Disclosures

Spine. 2023;48(16):1155-1165.

Abstract and Introduction

Abstract

Study Design: Systematic review and meta-analysis.

Objective: To perform a systematic review and meta-analysis of previous studies on HbA1c in preoperative risk stratification in patients undergoing spinal procedures and provide an overview of the consensus recommendations.

Summary of Background Data: Diabetes mellitus (DM) and hyperglycemia have been shown to be independent risk factors for increased surgical complications. Glycated Hemoglobin A1C (HbA1c), a surrogate for long term glycemic control, is an important preoperative parameter that may be optimized to reduce surgical complications and improve patient-reported outcomes. However, comprehensive systematic reviews on preoperative HbA1c and postoperative outcomes in spine surgery have been limited.

Methods: We systematically searched PubMed, EMBASE, Scopus, and Web-of-Science for English-language studies from inception through April 5^th, 2022, including references of eligible articles. The search was conducted according to PRISMA guidelines. Only studies in patients undergoing spine surgery with preoperative HbA1c values and postoperative outcomes available were included.

Results: A total of 22 articles (18 retrospective cohort studies, 4 prospective observational studies) were identified with level of evidence III or greater. The majority of studies (n=17) found that elevated preoperative HbA1c was associated with inferior outcomes or increased risk of complications. Random-effect meta-analysis demonstrated that patients with preoperative HbA1c >8.0% had increased risk(s) of postoperative complications (RR: 1.85, 95% CI: [1.48, 2.31], P<0.01) and that patients with surgical site infection (SSI) had higher preoperative HbA1c (Mean Difference: 1.49%, 95% CI: [0.11, 2.88], P=0.03).

Conclusion: The findings of this study suggest that HbA1c >8.0% is associated with an increased risk of complications. HbA1c was higher by 1.49% on average among patients with SSI when compared to patients who did not experience SSI. These results suggest that elevated HbA1c is associated with less favorable outcomes following spine surgery.

Level of Evidence: IV

Introduction

Diabetes mellitus (DM) is increasing in prevalence at an alarming rate and has become one of the major causes of morbidity and mortality worldwide, likely linked to an aging population with rising prevalence of obesity.^[1] In the United States, an estimated 37 million adults have diabetes and 96 million have prediabetes.^[2] Healthcare resource utilization and costs are twice as high in patients with diabetes compared to non-diabetic patients. Given the co-prevalence of DM and obesity, the incidence and prevalence of DM is expected to rise, making this a significant public health concern.^[3,4]

DM is a significant risk factor for complications following many forms of surgery. In patients with spinal disorders undergoing surgery, an estimated 25% have a history of DM.^[5–7] It is associated with poor convalescence after surgery and increases the incidence of wound infections, and adverse cardiovascular events. Perioperative short-term glycemic control has been shown to be associated with poor surgical outcomes in patients with and without DM, highlighting the role of stress hyperglycemia in this relationship.^[8,9]

Glycated hemoglobin A1C (HbA1c) has been used as a measure of diabetic control, reflecting glucose concentrations over the preceding months, and tight control is associated with reduced incidence of diabetic-related adverse events. The American Diabetes Association (ADA) released guidelines recommending that target HbA1c for patients with diabetes should be <7%, and furthermore that elective surgery should be avoided if the HbA1c exceeds 7%.^[10] However, it is unclear whether this cutoff is optimal for patients undergoing spine surgery. HbA1c measurement is currently not a standard part of the preoperative workup for patients undergoing spine surgery, nor is it specifically recommended, especially for non-diabetic patients or patients with undiagnosed DM.

While reasonably convincing data has shown that perioperative high glucose concentrations are associated with harm and that DM is associated with worse postoperative outcomes, it is not well established whether such a relationship exist for HbA1c and spine surgical outcomes. Therefore, the aim of this systematic review and meta-analysis is to clarify the relationship between preoperative HbA1c and postoperative outcomes following spine surgery.

1 2 3 4 5

Next Section

Abstract and Introduction
Methods
Results
Discussion
Conclusion
References
Sidebar

Table 1. Overview of All Included Studies
	Authors – Year	Study Design – Level of Evidence	Patients No./Female	Mean Age	Mean HbA1c	Elevated HbA1c Definition/Cutoff	Location No. patients (percentage)	Surgery Type	Levels of Surgery	Study findings
1	Roth et al – 2022¹⁶	Retrospective – IIb	408/172 (42.2%)	ACF: 55.8±10.1, PCLF: 63.6±9.2	ACF: 7.2±1.4, PCLF: 7.2±1.5	6.1, 6.8, 7.6	Cervical	ACF, PCLF	Multi-level	Positive
2	Wang et al – 2021¹⁵	Retrospective – IIb	705/304 (43.1%)	58.7±7.2, No-SSI: 58.4±7.1, SSI: 65±6.5	5.4±1.0, No-SSI: 5.3±0.8, SSI: 7.5±1.6	—	Lumbar	MIS-TLIF	—	Positive
3	Lim et al – 2021¹⁷	Retrospective – IIb	4778/2046 (42.8%)	HbA1c ≤ 8%: 61.5±13.4, HbA1c >8%: 62.0±11.9	—	8.0	Lumbar	Elective lumbar spine	Multi-level	Positive
4	Tanaka et al – 2021¹⁸	Prospective – IIb	440/202 (45.9%)	56.43±13.28	—	8.0	—	Spine	—	Positive
5	Ross et al – 2021⁸	Retrospective – IIb	174520/12422 (7.06%)	59±13	—	—	Cervical, thoracic, lumbar	Decompression alone, fusion and/or instrumentation Unspecified	—	Positive
6	Nagoshi et al – 2021²⁰	Retrospective – IIb	253/72 (28.5%)	DM: 61.9±9.1, non-DM: 59±9.4	DM: 6.4±1.1	—	Cervical	Laminoplasty, laminectomy, anterior/posterior fixation	—	Negative
7	Wiley et al – 2020²¹	Prospective – IIb	5444/3181 (58.4%)	54.20±17.12	—	7.0	Lumbar	Lumbar spine fusion	—	Positive
8	Hwang et al – 2019⁹	Retrospective – IIb	92/54 (58.7%)	SSI: 70.0±8.4, no SSI: 66.7±8.6	SSI: 6.78±1.29, no SSI: 5.99±0.77	6.9	Lumbar	PLF, PLIF	Single-level	Positive
9	Peng et al – 2019¹⁴	Retrospective – IIb	523/228 (43.6%)	—	—	7.5	Lumbar	Posterior lumbar surgery	—	Positive
10	Tanishima et al – 2019²²	Prospective – IIb	61/24 (39.3%)	Non-DM: 69.6±13.3, DM: 69.3±10.7	Non-DM: 5.6±0.4, DM: 6.9±0.9	—	Cervical	Laminoplasty, anterior decompression and fusion	—	Negative
11	Liow et al – 2018⁶	Retrospective – IIb	58/21 (36.2%)	DM: 59.1±10.4, non-DM: 59.2±10.2	—	7.0	Cervical	ACDF	Single-level	Negative
12	Cancienne et al – 2017²³	Retrospective – IIIb	5194/-	—	—	7.5	Lumbar	LD	Single-level	Positive
13	Kim et al – 2017²⁴	Prospective – IIb	119/68 (57.1%)	65.53±8.42, HbA1c <6.5: -, HbA1c ≥ 6.5: -	7.08±0.94, HbA1c <6.5: 6.39±0.07, HbA1c ≥ 6.5: 7.34±0.65	6.5	Lumbar	LD	Multi-level	Positive
14	Cancienne et al – 2017²⁵	Retrospective – IIIb	3341/-	—	—	7.5	Cervical	ACDF	—	Positive
15	Silverstein et al – 2016²⁶	Retrospective – IIb	212/90 (42.5%)	56±15, non-DM: -, DM: -	DM: 6.8±1.4	—	Lumbar	LD	—	Negative
16	Hikata et al - 2014²⁷	Retrospective – IIb	345/232 (67.2%)	54.7, HbA1c ≥ 7.0: 62.1, HbA1c <7.0: 68.1	HbA1c ≥ 7.0: 7.4 (7.0–8.1), HbA1c <7.0: 6.6 (6.1–6.9)	—	Thoracic, lumbar	PF, PLF, PLIF	Multi-level	Positive
17	Takahashi et al – 2013²⁸	Retrospective – IIIb	165/87 (52.7%)	DM: 70.9±7.2, non-DM 68.6±8.1	—	6.5	Lumbar	LD: fusion and non-fusion	—	Positive
18	Satake et al – 2013²⁹	Retrospective – IIb	110/-	SSI: 66.2±2.2, no SSI: 66.9±1.0	SSI: 6.88±0.3, no SSI: 6.80±0.1	—	Spine	Spinal instrumentation	—	Negative
19	Dokai et al – 2012³⁰	Retrospective – IIb	78/31 (39.7%)	DM: 72.0 (53–83), non-DM: 71.1 (35–87)	—	—	Cervical	ELAP	Multi-level	Positive
20	Wu et al – 2011³¹	Retrospective – IIb	126/54 (42.9%)	60.4±11.8, SL: 64.8±8.8, non-SL: 59.3±12.2	SL: 8.92±1.78, non-SL: 7.31±1.19	8.0	Lumbar	Lumbar dynamic stabilization	Multi-level	Positive
21	Walid et al – 2010³²	Retrospective – IIb	787/419 (53.2%)	54.5 (14–92)	—	6.1	Cervical, lumbar	LMD, ACDF, LDF	—	Positive
22	Walid et al – 2010³³	Retrospective – IIb	442/229 (51.8%)	No DM: 60±14, HbA1c: 59±11, DM: 69±9 No DM: 52±10, HbA1c: 58±9, DM: 60±10 No DM: 55±13, HbA1c: 54±10, DM: 59±7	—	6.1	Lumbar Cervical lumbar	LMD ACDF LDF	— — —	Positive

ACDF indicates Anterior cervical discectomy and fusion; ACF, Anterior cervical fusion; C, Complication group; DM, Diabetes Mellitus; ELAP, Expansive laminoplasty; LD, Lumbar decompression; LDF, Lumbar decompression and fusion; -, not reported; LMD, Lumbar microdiscectomy; MIS-TLIF, Minimally invasive transforaminal lumbar interbody fusion; -, Not reported; NC, No-complication group; PCLF, Posterior cervical laminectomy and fusion; PLF, Posterolateral fusion; PLIF, Posterior interbody fusion; SL, Screw loosening group; SSI, Surgical site infections.

Table 2. Included Studies Evaluating Preoperative HbA1c Levels and Postoperative Outcomes
	Authors – Year	HbA1c Cutoff	Reoperation Rate	Readmission Rate	Return to ER Rate	Complication Rate	LOS	Cost	SSI	RR
1	Roth et al – 2022¹⁶	ACF: 6.8 PCLF: 7.6	ACF: AUC=0.61, 95%CI 0.52–0.69, P=0.078	PCLF: AUC=0.63, 95%CI 0.50–0.75, P=0.185	—	—	—	—	—	—
2	Wang et al – 2021¹⁵	—	—	—	—	—	—	—	SSI vs non-SSI:Mean HbA1c: 7.5 vs 5.3, P<0.001HbA1c univariate: OR=11.394 (5.876–22.092), P<0.001; HbA1c multivariate: OR=36.125 (27.899–44.351), P<0.001	—
3	Lim et al – 2021¹⁷	8.0	—	OR 1.66, 95% CI 1.08–2.54; P=0.021	—	[OR] 1.81, 95% CI 1.20–2.73; P=0.005	OR 1.12,95% CI 1.03–1.23; P=0.009	—	SSI rate (OR 0.69, P=0.4668); patient satisfaction (OR 0.81, P=0.1967)	—
4	Tanaka et al – 2021¹⁸	8.0	—		2 (8.7%) P=0.0019	Bivariate OR=2.92; Multivariable OR=2.72	—	—	HbA1c<8.0% (P<0.0001)	—
5	Ross et al – 2021⁸	—	—	—	—	—	—	—	Presence of HbA1c test, OR=1.435 (1.367–1.505, P<0.0001); for each 1% HbA1c increase, OR=1.042 (1.017–1.068, P=0.0009)	—
6	Nagoshi et al – 2021²⁰	—	—	—	—	C5 palsy rate DM vs non-DM: 14.9% vs. 5.8%, P=0.04	—	—	—	—
7	Wiley et al – 2020²¹	7.0	—	90RA: χ2=11.76, P=0.008, OR=1.80 (1.16–2.65), P=0.007	90RER: χ2=14.08, P=0.0028, OR=1.5 (1.0–2.1), P=0.0458	—	—	—	—	—
8	Hwang et al – 2019⁹	6.9	—	—	—	—	—	—	ROC: P=0.003; AUC, 0.708; sensitivity, 62.5%; specificity, 70.6%Multivariate: P=0.008; OR=4.50; 95% CI, 1.49–13.62	—
9	Peng et al – 2019¹⁴	7.5	—	—	—	—	—	—	Multivariate: OR=4.629 (1.684, 10.349), P=0.01
10	Tanishima et al – 2019²²	—	—	—	—	—	—	—	—	Correlations between HbA1c and RR at 1 yr, CC=–0.02, P=0.99
11	Liow et al – 2018⁶	7.0	—	—	—	—	—	—	—	—
12	Cancienne et al – 2017²³	7.5	—	—	—	—	—	—	deep infection, OR=2.9, [1.8,4.9], P<0.0001; AUC=0.71, P=0.01, specificity=70%, sensitivity=53%	—
13	Kim et al – 2017²⁴	6.5	—	—	—	—	—	—	—	—
14	Cancienne et al – 2017²⁵	7.5	—	—	—	—	—	—	Deep infection, OR=1.84, [1.37,2.47]; P=0.008; AUC=0.67; P=0.022, specificity, 68%, sensitivity, 46%	—
15	Silverstein et al – 2016²⁶	—	—	—	—	—	—	—	—	—
16	Hikata et al - 2014²⁷	—	—	—	—	—	—	—	7.6 vs 6.9, P=0.006
17	Takahashi et al – 2013²⁸	6.5	—	—	—	—	—	—	—	—
18	Satake et al – 2013²⁹	—	—	—	—	—	—	—	6.88±0.3 vs 6.80±0.1, P=0.72	—
19	Dokai et al – 2012³⁰	—	—	—	—	—	—	—	—	HbA1c and RR at 6months: r=–0.718, P=0.006; final follow-up: r=–0.567, P=0.043
20	Wu et al – 2011³¹	8.0	—	—	—	SL vs non-SL: 8.92±1.78 vs 7.31±1.19, P=0.018 Loosening rate HbA1c >8 vs <8: 0.714 vs 0.286, P=0.021	—	—	—	—
21	Walid et al – 2010³²	6.1	—	—	—	—	—	LDF: $52499 vs $56176, P<0.01	—	—
22	Walid et al – 2010³³	6.1	—	—	—	—	LDF group: 6.96±5.897 vs 4.68±2.509, P<0.05	LDF group: $65974±18341 vs $54914±14034, P<0.05 ACDF group: $28363±7673 vs $36748±31970, P<0.05	—	—

ACDF indicates Anterior cervical discectomy and fusion; ACF, Anterior cervical fusion; C, Complication group; DM, Diabetes Mellitus; ELAP, Expansive laminoplasty; LD, Lumbar decompression; LDF, Lumbar decompression and fusion; LMD, Lumbar microdiscectomy; LOS, Length of stay; MIS-TLIF, Minimally invasive transforaminal lumbar interbody fusion; -, Not reported; MLRA, Multivariable logistic regression analysis; NC, No-complication group; PCLF, Posterior cervical laminectomy and fusion; PLF, Posterolateral fusion; PLIF, Posterior interbody fusion; RR, Recovery ratio = (postoperative JOA score-preoperative JOA score)/[17 (fullmark)-preoperative JOA score]; SL, Screw loosening group; SSI, Surgical site infections; ULRA, Univariable logistic regression analysis *100; -, not reported.

Table 3. Included Studies Evaluating Preoperative HbA1c Levels and Postoperative patient-reported Outcomes
	Authors – Year	HbA1c Cutoff	JOA	VAS	ODI	MCID	MCID	Others
1	Roth et al – 2022¹⁶	ACF: 6.1 PCLF: 6.8	—	—	—	—	For ACF, MCID NRS-Neck pain: 1.06 (0.84, 1.33), P=0.602; MCID NRS-Arm pain: 1.15 (0.91, 1.44), P=0.233; MCID NDI: 0.37 (0.22, 0.62), P<0.001. For PCLF, MCID NRS-Neck pain: 0.89 (0.63, 1.26), P=0.534; MCID NRS-Arm pain: 0.74 (0.52, 1.06), P=0.102; MCID NDI: 0.82 (0.58, 1.16), P=0.272	—
2	Wang et al – 2021¹⁵	—	—	—	—	—	—	—
3	Lim et al – 2021¹⁷	8.0	—	—	—	OR=0.72, [0.53, 0.98], P=0.034	—	—
4	Tanaka et al – 2021¹⁸	8.0	—	—	—	—	—	—
5	Ross et al – 2021⁸	—	—	—	—	—	—	—
6	Nagoshi et al– 2021²⁰	—	DM vs non-DM 13.9±2.7 vs 14.2±4.6, P=0.58 Preop HbA1c vs JOA: r=0.24, P=0.13; final followup HbA1c vs JOA: r=0.17, P=0.29	DM vs non-DM 23.3±28.6 vs 21.5±25.8, P=0.77	—	—	—	—
7	Wiley et al – 2020²¹	7.0	—	—	—	—	—	—
8	Hwang et al – 2019⁹	6.9	—	—	—	—	—	—
9	Peng et al – 2019¹⁴	7.5	—	—	—	—	—	—
10	Tanishima et al – 2019²²	—	—	—	—	—	—	—
11	Liow et al – 2018⁶	7.0	—	—	—	—	—	Functional outcomes, fulfilment of expectations, or patient satisfaction at 2 y P>0.05
12	Cancienne et al – 2017²³	7.5	—	—	—	—	—	—
13	Kim et al – 2017²⁴	6.5	Pearson coefficient 6 mo postop: 0.541 (P=0.009), 12 mo postop: 0.53 (P=0.012), 24 mo postop: 0.533 (P=0.01)		Pearson coefficient 6 mo postop: 0.513 (P=0.011), 12 mo postop: 0.485 (P=0.015), 24 mo postop: 0.505 (P=0.012)	—	—	—
14	Cancienne et al – 2017²⁵	7.5	—	—	—	—	—	—
15	Silverstein et al –2016²⁶	—	—	—	—	Improvements in EQ-5D (P=0.47), PDQ (P=0.46), PHQ-9 (P=0.69) exceeding the MCID	—	EQ-5D (P=0.13), PDQ (P=0.46), PHQ-9 (P=0.20)
16	Hikata et al – 2014²⁷	—	—	—	—	—	—	—
17	Takahashi et al – 2013²⁸	6.5	JOA: HbA1c≥6.5%, 1.26 (0.45, 3.5), N=41; HbA1c<6.5%, 1.30 (0.36, 4.65), N=124	VAS-LBP: HbA1c≥6.5%, 2.37 (0.99, 5.70), N=41; HbA1c<6.5%, 1.97 (0.63, 6.13), N=124 VAS-Leg Pain: HbA1c≥6.5%, 1.01 (0.35, 2.94), N=41; HbA1c<6.5%, 1.22 (0.36, 4.17), N=124 VAS-LBP: HbA1c≥6.5%, 1.62 (0.68, 3.81), N=41; HbA1c<6.5%, 1.56 (0.54, 4.53), N=124	—	—	—	Low back pain: OR=2.37; 95% CI, 0.99–5.70
18	Satake et al – 2013²⁹	—	—	—	—	—	—	—
19	Dokai et al – 2012³⁰	—	—	—	—	—	—	—
20	Wu et al – 2011³¹	8.0	—	—	—	—	—	—
21	Walid et al – 2010³²	6.1	—	—	—	—	—	—
22	Walid et al – 2010³³	6.1	—	—	—	—	—	—

ACDF indicates Anterior cervical discectomy and fusion; C, Complication group; DM, Diabetes Mellitus; ELAP, Expansive laminoplasty; EQ-5D, EuroQol 5-Dimensions; JOA, Japanese Orthopedic Association; LD, Lumbar decompression; LDF, Lumbar decompression and fusion; LMD, Lumbar microdiscectomy; MCID, Minimal clinically important differences; MIS-TLIF, Minimally invasive transforaminal lumbar interbody fusion; -, Not reported; NC, No-complication group; NDI, Neck disability index; PDQ, Pain Disability Questionnaire; PHQ-9, Patient Health Questionnaire; -, not reported.; PLF, Posterolateral fusion; PLIF, Posterior interbody fusion; SL, Screw loosening group; SSI, Surgical site infections; VAS, Visual analog scale.