Figure 1.
Schematic of the density-based tests to identify SCD.
Both versions of the SCD-AMPS are designed to separate dense red blood cells present in SCD from whole blood. Blood passes through the phases—top (T) and bottom (B) for SCD-AMPS-2 and top (T), middle (M), and bottom (B) for SCD-AMPS-3—upon centrifugation. If sickled cells are present, they collect at the interface between the bottom phase and the seal (B/S), and provide a visual readout for the presence of SCD. In SCD-AMPS-3, the additional phase allows the discrimination of Hb SS from Hb SC by evaluating the distribution of red cells at the upper interfaces (between the top and middle phases (T/M) and the middle and bottom phases (M/B).
Table 1.
Inclusion and Exclusion Criteria for Study.
Table 2.
Basic Characteristics of the Study Population.
Figure 2.
Schematic representation of the five levels of evaluation of the bottom of an SCD-AMPS test.
Levels 1 through 5 correspond to increasing levels of accumulation of red cells at the interface between the bottom phase and the seal (B/S) compared to the accumulation of red cells at the interface between the top and bottom phase (T/B).
Figure 3.
Receiver Operating Characteristic (ROC) curves of SCD-AMPS-2 and SCD-AMPS-3 including all data from Zambia shows fair discriminative ability.
The amount of red cells below the bottom phase of each test was evaluated by eye and classified on a five point scale. Both tests showed an ability to discriminate sickle cell disease from non-disease. In general, the specificity was lower than the sensitivity of the tests.
Table 3.
Tabulation of Results of SCD-AMPS Tests Compared to Reference Test Results by Hemoglobin Electrophoresis.
Figure 4.
The sensitivity and specificity of SCD-AMPS as a function of the amount of time between collecting samples and running tests.
The specificity shows a decline over each 24 hour increment, with a significant decline over 48 hours (p-value <0.0005). The sensitivity increased between the first and second time interval, but then decreased between the second and third interval (p-values <0.01). The sample size used for each time interval is provided below each bar.
Table 4.
Sensitivity of SCD-AMPS tests on specific subpopulations.
Figure 5.
Equipment for the SCD-AMPS rapid test.
All the equipment necessary to run the rapid test in a rural clinic fits inside a backpack and were evaluated at rural health centers in Zambia.