Microdialysis Sampling Principles
Microdialysis is founded on the principle of passive diffusion across a semi-permeable membrane implanted in living tissue. A perfusion fluid, chosen to match the ionic composition of the surrounding extracellular fluid, is continuously pumped through the probe lumen at a controlled flow rate, typically between 0.1 and 5.0 microliters per minute. Small molecules in the extracellular space diffuse across the membrane into the perfusate along their concentration gradient, while the continuous flow carries these molecules to an outlet tube where they are collected as discrete fractions. The technique was pioneered by Urban Ungerstedt in the 1970s and has since become the gold standard for in vivo neurochemical monitoring in freely moving animals. The key analytical parameter is relative recovery — the ratio of dialysate concentration to true extracellular concentration. Relative recovery is governed by the Bungay mass transport model, which accounts for resistances at the membrane, the extracellular space, and the probe-tissue interface. Recovery depends on flow rate (inversely), membrane area (directly), membrane material and MWCO, analyte diffusion coefficient, and tissue tortuosity. At typical experimental flow rates of 0.5-2.0 microliters per minute, relative recovery for small neurotransmitters ranges from 5-25%, meaning the dialysate represents a diluted but proportional sample of the extracellular milieu. This dilution factor must be considered when interpreting absolute concentrations, although most microdialysis studies report relative changes from baseline, making exact recovery less critical. The no-net-flux and retrodialysis calibration methods can be used to estimate in vivo recovery when absolute quantification is required.
Optimizing Flow Rate and Fraction Volume
The choice of flow rate and fraction collection interval represents the central trade-off in microdialysis experimental design: temporal resolution versus analytical sensitivity. Higher flow rates produce larger fraction volumes in a given time interval, providing more sample for analytical detection, but reduce relative recovery because the perfusate equilibrates less with the tissue. Lower flow rates improve recovery but yield smaller volumes that may fall below the analytical method's limit of detection or minimum injection volume. The fraction volume is simply the product of flow rate and collection interval: for example, 1.0 microliter per minute collected for 15 minutes yields a 15-microliter fraction. The minimum usable fraction volume is determined by the analytical platform — HPLC-ECD systems typically require 5-20 microliters for injection, while LC-MS/MS methods can work with as little as 1-5 microliters. When designing a collection schedule, researchers must also consider the number of fractions needed: a 3-hour experiment with 10-minute fractions generates 18 fractions per probe, while 5-minute fractions double that to 36 fractions, increasing analytical workload and consumables cost. For pharmacokinetic studies where drug concentrations change slowly over hours, 15-20 minute fractions are adequate. For fast neurotransmitter dynamics such as phasic dopamine release during operant behavior, shorter fractions (1-5 minutes) or even online detection systems may be necessary. A practical approach is to use longer baseline fractions (15-20 minutes) before the experimental manipulation, then switch to shorter fractions (5-10 minutes) during the period of interest, and return to longer fractions during the recovery phase. This mixed-interval protocol maximizes temporal resolution when it matters most while keeping total fraction count manageable.
Dead Volume and Temporal Resolution
Dead volume — the internal volume of the outlet tubing between the probe tip and the fraction collector — introduces a fixed time delay between the moment an analyte crosses the membrane and when it appears in the collected fraction. This delay, calculated as dead volume divided by flow rate, must be subtracted from fraction timestamps to accurately align neurochemical data with experimental events. For a typical setup with 30 cm of polyethylene PE-10 tubing (0.28 mm ID), the dead volume is approximately 18.5 microliters, creating an 18.5-minute lag at 1.0 microliter per minute — a delay that can be larger than the fraction interval itself, causing collected samples to reflect events from the previous collection period. Reducing dead volume requires using shorter outlet tubing, narrower-bore tubing (e.g., fused silica with 50-75 micrometer ID), or both. However, narrower tubing dramatically increases backpressure (fourth-power relationship with diameter), which can exceed the syringe pump's pressure limit or cause probe membrane failure. The optimal outlet tubing configuration balances dead volume against practical backpressure constraints. Fused silica capillary (75 micrometer ID) at 20 cm length gives approximately 0.88 microliters dead volume and a lag of only 53 seconds at 1.0 microliter per minute, but generates substantially more backpressure than standard PE tubing. Dead volume also affects temporal smearing: analyte pulses are dispersed by Taylor dispersion as they travel through the tubing, broadening sharp concentration transients into smoother peaks. This dispersion is minimized in narrow-bore tubing at higher flow rates. When planning an experiment, always calculate the dead volume delay and verify that it does not exceed your desired temporal resolution. If the delay is unacceptable, shorten the tubing, reduce the ID, or increase the flow rate (accepting the recovery trade-off).
