The aim of manual infusion techniques is same as that of target controlled. We aim at a plasma therapeutic concentration, assisted by mental mathematics to decide the initial bolus and the subsequent infusion rates for a desired target concentration (and clinical effect). Analogous to inhalational anesthesia, it requires experience to choose the appropriate loading doses and infusion rates for different patients and surgical procedures. Addition of any drugs (some of them are a necessity, e.g., opioid either remifentanil or alfentanil) adds another layer of complexity. The drug infusion rates and the need for any additional boluses is usually guided by the intensity of surgical stimulation and accompanying hemodynamic response.

Irrespective of the infusion technique chosen, it is important to bear in mind that technical problems like unnoticed IV (intravenous) line disconnection or accidental infiltration into interstitial or subcutaneous space can lead to awareness. In the commonest propofol-remifentanil-based TIVA, tachycardia and hypertension (or even movement in a non-paralyzed patient) due to loss of remifentanil effect might be seen first. This is due to the shorter and predictable context-sensitive halftime of remifentanil. Depending on the duration of infusion, it can take 10–20 min for the effects of propofol to wear off [14]. A gas can be introduced if a disconnection or infiltration is noticed, while efforts are being made to secure another intravenous access.

At the outset, any contraindications to TIVA must be ruled out. Allergy to propofol or any of the components will obviate its use. A strong opioid like remifentanil, sufentanil, or alfentanil is almost mandatory especially if muscle relaxants are to be avoided. An IV cannula is essential to initiate and maintain TIVA. As a result children and adults who need inhalational agents to insert IV are not candidates for TIVA although, following inhalation induction, one can switch over to intravenous agents (after securing IV access). Excessive preoperative anxiety warrants need of a sedative anxiolytic (e.g., midazolam), although this may prolong wake-up times after termination of the infusions. The bolus dose is dictated by the lean body mass, age, sex, and comorbidity. With 1.5–2 μg/kg of remifentanil over 2 min, one normally would require 2–2.5 mg/kg (bolus over 2 min) of propofol. With this dose, intubation can be achieved without relaxants or placement of a LMA can be done in 1–2 min after end of bolus dose. If TCI system is being used, target an initial propofol concentration of 5–6 μg/ml and remifentanil of 4–5 ng/ml [15]. Time to peak effect of propofol is around 2 min. It is important to bear in mind that there could be substantial bradycardia and hypotension at induction. Frequently preempting this with a small dose of pressor (like ephedrine) may be required [16]. It is especially advised in the young/healthy/ASA I who have predilection to bradycardia (due to heightened vagal tone) and very elderly who often show hypotension. The latter is because of heightened pharmacodynamic response to both propofol and remifentanil with insufficient sympathetic compensation [17].

To compensate for the rapid fall in the plasma propofol concentration after initial bolus (mainly due to redistribution), higher infusion rates are required in the first 10–15 min of anesthesia. However, due consideration should be given to prolonged periods of absent stimulation depending on the surgeon, surgery, and the country practices. In USA typically these times are 10–15 min. One would be tempted to cut the infusion rates during this period; however, it is important to supplement bolus 1–2 min before incision to bring up the plasma concentrations. Otherwise, there could be patient movement that would be perceived as awareness by the surgeons and nurses. However, the incidence of awareness is rare with intravenous anesthesia, barring technical issues or equipment malfunction [18]. After the initial surgical stimulation, depending on the surgical procedure and the hemodynamic responses, infusion rates (or target concentrations) can be titrated accordingly.

Context-sensitive halftime is defined as the time required for drug plasma concentration to decrease by 50 % after cessation of the drug infusion. The context here is the duration for which the drug infusion is run. The timing of the cessation of infusion (at the end of surgery) should be guided by the context-sensitive halftime of the drug in question. As remifentanil washout is context insensitive, due to rapid metabolism, minimal redistribution, and inactive metabolites, its plasma half-life is typically 2–3 min irrespective of the duration of infusion. However, with propofol, there is prolongation of the 50 % plasma decrement times after prolonged infusions, and this has to be taken into consideration while reducing and stopping its infusion. It is also known that elimination half-life of a drug often fails to predict the context-sensitive halftime as factors other than elimination half-life (redistribution, active metabolites, etc.) also play a significant role in the fall of drug concentration post infusion. For example, the context-sensitive halftime for propofol is significantly shorter than midazolam, despite the fact that it has much longer elimination half-life (around 400 min versus 170 min for midazolam) [19]. Clinical utility of this translates into the fact that concerns of delayed respiratory depression using intravenous opioids must not be guided by longer half-lives rather by context-sensitive halftime. Clinical results prove that despite much longer half-life of sufentanil when compared to alfentanil, the context-sensitive halftime of sufentanil is much shorter and thus alfentanil is more liable to cause delayed respiratory depression despite shorter elimination half-life [20]. If significant postoperative pain is expected, it is important to address this either with nerve block or alternate analgesic before turning off remifentanil infusion because of shorter context-sensitive halftime.

Although halftimes predict time for drop in the plasma concentrations by half, they do not necessarily reflect “wake-up” times. Most patients respond to verbal commands even at concentrations close to half of the plasma anesthetic concentrations. However, some patients may wake up at concentrations more than 50 % of anesthetic concentrations, and others may not wake up at concentrations even significantly below 50 %. Coadministration of any benzodiazepines and opioids with longer context-sensitive halftime is another factor affecting wake-up times. Although it is more practical to administer opioids with shorter context-sensitive halftimes (like remifentanil) especially for long cases, opioids with long context-sensitive halftimes (sufentanil, alfentanil) can be administered intelligently to achieve shorter recovery times especially in surgeries with long skin closure times. A good knowledge of pharmacokinetics is essential. Empirically the infusion rate for short periods after initial bolus is approximately the product of elimination clearance and target plasma concentration [19].

Enormous variability in the pharmacokinetics and pharmacodynamics among intravenous anesthetic agents would mean that no model can accurately predict the plasma concentrations. Even if there is a way of accurately estimating or measuring the plasma levels, the variable response needs to be addressed. Pharmacodynamic variability can be anticipated and allowance made in some situations like elderly or existing narcotic use (or abuse). But there are many situations like changing surgical stimulation that cannot be always anticipated or quantified accurately. Anesthesia providers try to match the depth of anesthesia with the intensity of surgical stimulation. Sometimes it is impossible to find any factors to explain very high propofol/remifentanil concentrations required to maintain acceptable hemodynamic parameters or depth of anesthesia as measured by EEG (electroencephalography)-based monitoring. Apparent lighter levels as evidenced by patient movement may punctuate a perfectly smooth conduct of anesthesia. This could be annoying and alarming to the surgeon and may even disrupt the surgical procedure. BIS (bispectral index) may be capable of reflecting this; however, unfortunately only after the event has occurred. Fortunately these events are almost never reported as awareness in the absence of technical administration problems. These problems cannot be adequately addressed by a model of any complexity or microprocessor of any speed.

Advantages of using TCI pumps for the administration of TIVA are

A question frequently asked is “Can TIVA be practiced safely without the use of remifentanil as an analgesic?” The answer is a qualified yes. Avoiding remifentanil increases use of muscle relaxant and drugs for controlling hemodynamic responses. Remifentanil in the clinical doses is an excellent agent to produce controlled hypotension. Any analgesic other than remifentanil requires the use of additional muscle relaxants to avoid patient movement and facilitate intubation. For example, although, remifentanil is almost equipotent to fentanyl, the infusion rates required similar for clinical effect for fentanyl are four to five times (“cardiac” anesthetic doses) higher [21]. These higher doses, if run for longer durations, result in drug accumulation, thus leading to residual or prolonged effects. Moreover, fentanyl is context sensitive unlike remifentanil, with significant halftime prolongation at the cessation of intermediate to long periods of infusion. Fentanyl might be an acceptable choice for short procedures [22]. However, if large component of analgesia is provided using neuroaxial blockade or plexus/peripheral nerve blockade, then one can safely perform TIVA with smaller infusion doses of remifentanil or fentanyl. None of the available opioids allow rapid titration like remifentanil which makes it the drug of choice for procedures like awake fiberoptic intubation (with or without propofol) and many endoscopy procedures [23]. The reader is referred to [24] for an excellent review on remifentanil.