Introduction
Methods
-
To identify the estimated prevalence, anatomical distribution and the tissue morphology of LH with US and/or palpation assessment
-
To identify factors associated with LH in US assessed cases
-
To estimate the sensitivity of palpation in LH detection, with US as the reference
Search Strategy
Selection Criteria
Data Extraction
Data Synthesis
Quality Assessment
Ethical Considerations
Results
Author (year), country | Study aim(s) | Sample population | Insulin therapy* | Detection method(s), assessor(s), region(s) | Outcome measure(s) |
---|---|---|---|---|---|
RCT | |||||
Famulla et al. (2016) [4] Germany | Determine the impact of LH on insulin activity | Sample size: 13 Population: T1DM Mean age (years, SD): 50.1 ± 10.5 Gender (M/F): ND Mean diabetes duration (years, SD): 26.8 ± 6.9 | Insulin needle injection | Detection Observation, palpation (pre-defined examination procedure) and US Assessors 2 independent investigators Region Abdomen | LH detection, insulin absorption in LH regions, diabetes outcomes: insulin and glucose levels |
Observational studies | |||||
Bertuzzi et al. (2017) [19] Italy | Characterisation of LH using US | Sample size: 20 Population: T1DM Mean age (years, SD): 37 ± 12 Gender (M/F): 8/12 Mean diabetes duration (years, SD): 22 ± 12 | CSII, MDI | Detection Palpation (pre-defined criteria), US (pre-defined criteria) Assessors Investigators Region Abdomen, arm, gluteus | LH detection, metabolic parameters: HbA1c |
Blanco et al. (2013) [20] Spain | Assess LH prevalence and determine its correlation with clinical and public-health factors | Sample size: 430 Population: T1DM (41%) + T2DM (59%) Mean age (years, SD): 49 ± 22.8 Gender (M/F): 221/202 Diabetes duration (years, range) 6–15 | Insulin pen | Detection US (pre-defined characteristics) Assessors CD Region Abdomen | Prevalence of LH, risk factors of LH, diabetes outcomes: hypoglycaemia, glucose variability |
Conwell et al. (2008) [21] Canada | Describe dermatological changes with CSII therapy | Sample size: 50 Population: T1DM Mean age (years, SD): 13.3 ± 3.5 Gender (M/F): 24/26 Mean diabetes duration (years, SD): 6.5 ± 3.7 | CSII | Detection Palpation (pre-defined criteria) in all 50 patients, US (pre-defined criteria) in 8 out of 50 patients Assessors 1 trained investigator Region Abdomen, back/buttock, legs | Frequency of LH, dermatological changes including LH |
Davidenko et al. (2014) [22]—abstract, Russia | Develop estimation risk model of insulin induced LH | Sample size: 140 Population: Diabetes (NS) Mean age (years, SD): ND Gender (M/F): 51/89 Mean diabetes duration (years, SD): ND | NS | Detection Observation, palpation, US Assessors CD Region ND | Frequency of LH, estimation of insulin as a risk factor of LH |
Gentile et al. (2016) [14] Italy | Identify inexpensive and accurate LH detection method | Sample size: 40 Population: Diabetes (NS) Mean age (years, SD): 54 ± 15 Gender (M/F): 16/24 Mean diabetes duration (years, SD): ND | Insulin needle injection | Detection Palpation (pre-defined criteria), US (pre-defined criteria) Assessors Trained vs non-trained nurses Region Abdomen, arm, thigh | Prevalence of LH, detection accuracy of LH between trained and non-trained HPs |
Kapeluto et al. (2015) [33]—abstract, Canada | Establish a criteria for LH detection using US | Sample size: 7 Population: Insulin-users Mean age (years, SD): ND Gender (M/F): 6/1 Mean diabetes duration (years, SD): ND | NS | Detection Palpation, US Assessors Single team + radiologist Region Abdomen | LH diagnostic criteria using US |
Kasperska-Czyzyk et al. (2000) [23]—abstract, Poland | Determine the robustness of US in the diagnosis and characterisation of LH | Sample size: 30 Population: T1DM (33%) + T2DM (60%) + secondary diabetes (7%) Mean age (years, SEM): 60 ± 2 Gender (M/F): 16/14 Mean diabetes duration (years, SD): ND | Insulin needle injection | Detection Palpation, US Assessors CD Region Abdomen, arms, thighs | Detection accuracy of LH between US and palpation |
Mulnier et al. (2017) [24]—abstract, UK | Evaluate the feasibility of US to detect and characterise LH | Sample size: 26 Population: T1DM Mean age (years): 41 Gender (M/F): ND Mean diabetes duration (years): 22.7 | Insulin needle injection | Detection: US Assessors: CD Region: CD | LH detection, changes in subcutaneous tissue with insulin use |
Nasser et al. (2017) [25] Bahrain | Determine LH prevalence and risk factors of LH | Sample size: 95 Population: T1DM (3%) + T2DM (97%) Age (years, range): 40 to ≥ 70 Gender (M/F): 23/72 Diabetes duration (years, range): 5 to ≥ 20 | Insulin needle injection | Detection: US Assessors: Clinical nurse Region: Abdomen, arm, thigh | Prevalence of LH, risk factors of LH |
Patrakeeva et al. (2014) [26]—Abstract, Russia | Evaluate LH frequency using different insulin regimens and risk factors of LH | Sample size: 29 Population: T1DM Mean age (years, SD): 27 ± 4 Gender (M/F): ND Mean diabetes duration n (years, SD): 13.7 ± 2.1 | CSII, MDI | Detection: US Assessors: ND Region: ND | Prevalence of LH, risk factors of LH |
Perciun (2010) [27] Romania | Characterise LH sites using US | Sample size: 40 Population: Diabetes (NS) Age (years, range): 15 – 65 Gender (M/F): 14/26 Mean diabetes duration (years, SD): ND | Insulin needle injection | Detection palpation (pre-specified criteria), US (pre-specified criteria) Assessors ND Region Abdomen, arm, thigh, buttocks | Diagnosis and evaluation of LH sites |
Perciun et al. (2014) [28] Romania | Compare the diagnosis of dystrophies between US and palpation | Sample size: 53 Population: T1DM Age (years, range): 2–15 Gender (M/F): 33/20 Diabetes duration (years, range): 1–13 | Insulin needle injection | Detection Palpation (pre-specified criteria), US (pre-specified criteria) Assessors CD Region Abdomen, arm, thigh, buttocks | Frequency of hypertrophic sites |
Volkova et al. (2013) [29]—abstract, Russia | Compare the frequency of LH between US and palpation | Sample size: 215 Population: Diabetes (NS) Mean age (years): 46 Gender (M/F): 142/73 Mean diabetes duration (years, SD): ND | Insulin needle injection | Detection: Observation, palpation, US Assessors: ND Region: Paraumbilical/buttocks, hips, shoulders | Detection accuracy of LH between US and palpation |
Wang et al. (2014 ) [30]—abstract, China | Estimation of skin and subcutaneous layer thickness and prevalence of LH | Sample size: 509 Population: T1DM + T2DM (% NS) Age (years, range): 18–85 Gender (M/F): ND Diabetes duration (years, SD): ≥ 1 | Insulin needle injection | Detection US Assessors ND Region Abdomen, arm, thigh, buttocks | Prevalence of LH, estimation risk of subcutaneous and intramuscular injection |
Case reports | |||||
Blanco et al. (2013) [20] Spain | Assess diabetes history of LH patient | Sample size: 1 Population: T1DM Age (years): 32 Gender: M Diabetes duration (years): 18 | Insulin needle injection | Detection Palpation, US Assessors ND Region Abdomen | Change in LH, diabetes-related outcomes: HbA1c, insulin use, hypoglycaemic episodes |
Perciun et al. (2012) [32] Romania | Analyse cutaneous and subcutaneous dystrophies using two different ultrasounds | Patient 1 Population: Diabetes (NS) Age (years): 55 Gender: M Diabetes duration (years): ND | Insulin needle injection | Detection Physical examination, US Assessors ND Region Abdomen | Presence of cutaneous damage, metabolic control |
Patient 2 Population: Diabetes (NS) Age (years): 30 Gender: F Diabetes duration (years): ND | CSII | Detection US Assessors ND Region ND |
Quality Appraisal
LH Prevalence, Anatomical Distribution and Tissue Morphology
Author (year) | Insulin therapya | LH prevalence (%) based on detection method | ||
---|---|---|---|---|
US | Palpation | Not specified | ||
Blanco et al. (2013) [20] | Pen | 64.4 | ||
Conwell et al. (2008) [21] | CSII | 44.0 | ||
Davidenko et al. (2014) [22] | NS | 84.0 | ||
Gentile et al. (2016) [14] | Needle injection | 48.8 | ||
Nasser et al. (2017) [25] | Needle injection | 36.8 | ||
Patrakeeva et al. (2014) [26] | CSII, MDI | 76.0 | ||
Perciun (2010) [27] | Needle injection | 88.0 | ||
Volkova et al. (2013) [29] | Needle injection | 86.5 | 37.0 | |
Wang et al. (2014) [30] | Needle injection | 14.5 |
Author (year) | Insulin therapya | Anatomical distribution of LH | |||
---|---|---|---|---|---|
Abdomen | Arm | Thigh/gluteus | Other | ||
Bertuzzi et al. (2017) [19] | CSII, MDI | 100% (US, Palp.) | 25.0% (US), 20.0% (Palp.) | 25.0% (US), 20.0% (Palp.) | |
Blanco et al. (2013) [20] | Pen | NDb | |||
Conwell et al. (2008) [21] | CSII | SS: 6.1 ± 3.3 | SS: 4.9 ± 4.2 | ||
Gentile et al. (2016) [14] | Needle injection | 40.0% | 35.0% | 25.0% | |
Kasperska-Czyzyk et al. (2000) [23] | Needle injection | 61.8% | 17.6% | 20.6% | |
Nasser et al. (2017) [25] | Needle injection | S/SC: 3.1/23.3 mm | S/SC: 2.9/12.3 mm | S/SC: 3.2/12.3 mm | |
Volkova et al. (2013) [29] | Needle injection | Paraumbilical/buttocks: 61.0% Paraumbilical/buttocks + hips: 15.0% Paraumbilical/buttocks + shoulders: 11.0% |
Characterisation of LH Using US
Risk Factors of LH
Author (year) | Risk factors of LH | Statistical value |
---|---|---|
Blanco et al. (2013) [20] | Needle reuse Patient-reported injection site rotation Nurse-reported injection site rotation Nurse observed + patient claimed injection site rotation | P = 0.008 P = 0.001 P = 0.0001 P = 0.0001 |
Conwell et al. (2008) [21] | BMI z-score 0.60 ± 0.76 Needle insertion angle at 90º | r = − 0.3, P = 0.04 P = 0.03 |
Davidenko et al. (2014) [22] | Insulin use | AUC > 0.5–86.0% |
Nasser et al. (2017) [25] | Level of education Number of injections Injection site: Arm Injection site: Abdomen | P = 0.02 P = 0.02 P = 0.04 P = 0.001 |
Patrakeeva et al. (2014) [26] | Glucose variability Incorrect insulin injection technique/infusion set changing | r = 0.8 r = 0.7 |