Study selection and characteristics
A total of 482 articles were identified from the databases searched. One hundred and seventeen duplicates were removed and 264 studies were excluded through an initial screening. After a full text assessment for eligibility of the remaining 101 articles, 19 studies were identified for inclusion in this meta-analysis. No additional studies were found through reference screening of the included papers. Figure 1 shows the flow of the literature search and study selection process. The included 19 studies containing a total of 708 patients with diagnosed or suspected insulinoma, provided relevant detection results of PET/CT, SPECT/CT, CT and MRI in those participants. These articles were published from 1996 to 2020. Table 1 showed details of the studies included.
Table 1
Characteristics of studies included
Study | Year of publication | Country | No. of patients | Mean Age, y (SD) | Male, No. (%) | Type of radiotracer | Patient selection |
Zimmer | 1996 | Germany | 10 | 58.4(14.0) | 20 | 111In-pentetreotide | Patients with suspected insulinoma |
Proye | 1998 | France | 10 | 46(2.7) | 33 | 111In-pentetreotide | Patients with confirmed insulinoma |
Schillaci | 2000 | Italy | 14 | 46.5(15.5) | 57.1 | 111In-pentetreotide | Patients with suspected insulinoma |
Christ | 2013 | Germany, Switzerland, and the UK | 25 | 55 | 37 | ¹¹¹In-DTPA-exendin-4 | Patients with suspected insulinoma |
Antwi | 2015 | Switzerland | 5 | 54.0 (10.6) | 40 | 68Ga-DOTA-exendin-4, 111In-DOTA-exendin-4 | Patients with suspected insulinoma |
Luo | 2016 | China | 52 | NR | 44.2 | 68Ga-NOTA-exendin-4 | Patients with confirmed insulinoma |
Prasad | 2016 | Germany | 13 | 53.1 | 30.8 | 68Ga-DOTATATE/DOTATOC | Patients with confirmed insulinoma |
Sharma | 2016 | India | 35 | 38.4(16.5) | NR | 68Ga-DOTANOC | Patients with suspected insulinoma |
Sowa-Staszczak | 2016 | Poland | 25 | 65.5(12) | 37.5 | 99mTc-GLP1 | Patients with suspected insulinoma |
Wei | 2016 | China | 33 | 48 | 30.3 | NR | Patients with confirmed insulinoma |
Nockel | 2017 | USA | 31 | 57.5 | 35.5 | 68Ga-DOTATATE | Patients with confirmed insulinoma |
Zhu | 2017 | China | 70 | 46 | 47.1 | NR | Patients with suspected insulinoma |
Antwi | 2018 | Switzerland | 52 | 49 | 23 | 68Ga-DOTA-exendin-4 | Patients with suspected insulinoma |
Luo | 2018 | China | 69 | 43.2 | 53.6 | 68Ga-NOTA-exendin-4 | Patients with suspected insulinoma |
Nakuz | 2018 | Austria | 10 | 53 | 20 | 18F-FDOPA | Patients with confirmed insulinoma |
Leroy-Freschini | 2019 | France | 24 | 52.5(14.2) | 25 | 18F-FDOPA | Patients with suspected insulinoma |
Pallvi | 2019 | India | 8 | 30 | 62.5 | 68Ga-DOTA-exendin-4 | Patients with suspected insulinoma |
Garg | 2020 | India | 14 | 34.1(15.8) | 50 | 68Ga-DOTATATE, 68Ga-NODAGA-exendin-4 | Patients with suspected insulinoma |
Kalff | 2020 | Australia | 24 | 52.6(19.2) | 54.2 | 68Ga-DOTA-Exendin-4 | Patients with suspected insulinoma |
NR, not reported. |
Quality Of Studies
Quality assessment by QUADAS-2 scale showed that 13 studies had low risk of bias for patient selection, 1 study had high risk of bias and 5 studies had unclear risk. 13 studies had low risk of bias for index test, risk of bias for others were unclear. Twelve studies had low risk of bias for reference standard, 1 study had high risk of bias, and 6 had unclear risk of bias. In the part of flow and timing, 3 studies had unclear risk of bias, 2 had high risk and 14 had low risk. Clinical applicability concerns of each study included were also evaluated (Fig. 2).
Diagnostic Performances Of Imaging Modalities
Eight studies reported outcomes of PET/CT imaging in the detection of insulinoma. They demonstrated a pooled sensitivity of 0.79 (95% CI: 0.54–0.92, I2 = 91.5%, p < 0.01) and a pooled specificity of 0.84 (95% CI: 0.20–0.99, I2 = 76.3%, p < 0.01). The pooled sensitivity and specificity of SPECT/CT were 0.77 (95% CI: 0.46–0.93, I2 = 87.4%, p < 0.01) and 0.45 (95% CI: 0.22–0.70, I2 = 0, p = 0.79). CT showed an overall sensitivity of 0.54 (95% CI:0.35–0.72, I2 = 80.2%, p < 0.01) and specificity of 0.75 (95% CI: 0.54–0.88, I2 = 0, p = 0.48). The pooled sensitivity and specificity for MRI were 0.54 (95% CI: 0.31–0.75, I2 = 75.0%, p < 0.01) and 0.65 (95% CI: 0.39–0.84, I2 = 0, p = 0.53), respectively (Fig. 3). The AUC values for PET, SPECT, CT and MRI were 0.86 (95% CI: 0.83–0.89), 0.49 (95% CI: 0.44–0.53), 0.74 (95% CI: 0.70–0.78) and 0.65 (95% CI: 0.60–0.69) (Fig. 4).
The concordance rates of PET, SPECT, CT, and MRI were 78% (95% CI: 66%-90%, I2 = 75.0%, p < 0.01), 74% (95% CI: 52%-97%, I2 = 92.8%, p < 0.01), 56% (95% CI: 41%-72%, I2 = 85.1%, p < 0.01), and 53% (95% CI: 33%-73%, I2 = 88.4%, p < 0.01) (Fig. 5).
Subgroup Analysis Of Molecular Imaging Modalities
The number of studies on GLP-1R or GLP-1R analog-based PET/CT was 7, the pooled sensitivity and specificity were 0.87 (95% CI: 0.62–0.97, I2 = 87.5%, p < 0.01) and 0.94 (95% CI: 0.24–1.00, I2 = 75.6%, p < 0.01). Two studies reported the results of somatostatin receptor-based PET/CT in the detection of insulinoma, the overall sensitivity and specificity were 0.39 (95% CI: 0.24–0.56, I2 = 89.4%, p < 0.01) and 0.14 (95% CI: 0–0.58, I2 = 19.5%, p = 0.265). With regard to SPECT/CT, the pooled sensitivity and specificity for GLP-1R or GLP-1R analog-based imaging were 0.83 (95% CI: 0.73–0.91, I2 = 79.8%, p = 0.007) and 0.40 (95% CI: 0.12–0.74, I2 = 0, p = 0.394), respectively. For somatostatin receptor-based scanning, the pooled sensitivity and specificity were 0.53 (95% CI: 0.36–0.69, I2 = 87.6%, p < 0.001) and 0.50 (95% CI: 0.12–0.88, I2 = 0, p = 0.962) (Table 2).
Table 2
Results of subgroup analysis of molecular imaging modalities
Modalities | Sensitivity | Specificity | +LR | -LR | DOR | AUC |
PET/CT | 0.79 | 0.84 | 5 | 0.25 | 20 | 0.86 |
GLP-1R analog-based | 0.87 | 0.94 | 15 | 0.14 | 111 | 0.95 |
SSTR-based | 0.39 | 0.14 | - | - | - | - |
SPECT/CT | 0.77 | 0.45 | 1.4 | 0.51 | 3 | 0.49 |
GLP-1R analog-based | 0.83 | 0.4 | - | - | - | - |
SSTR-based | 0.53 | 0.5 | - | - | - | - |
Publication Bias
Deeks’ funnel plot asymmetry tests yielded p-values of 0.02, 0.36, < 0.01 and 0.09 for PET/CT, SPECT/CT, CT and MRI, respectively.