Spinal anesthesia can provide anesthesia, analgesia, and muscle relaxation. It has been successfully employed for procedures involving the abdomen, perineum, lower extremities, and urogenital tract as well as obstetric surgeries [1, 2].

Palpation of the neck reveals a prominent spinous process that belongs to the 7th cervical vertebra, called vertebra prominens. Palpation of the scapula reveals the root of the spine of the scapula commonly corresponding to a T3 level. The lower border of the scapula ends in a tip that usually is at the T7 level. At the lumbar spine level, a line drawn between the top of the bilateral iliac crests typically crosses the body of the 4th lumbar vertebra or the L4/5 interspace. This line is called Tuffier’s line. While those landmarks are considered “traditional teaching,”, studies have highlighted that their use frequently results in inaccurate determination of the intervertebral space, yet at a degree that still allows safe neuraxial placement in clinical practice. For patients whose anatomical landmarks are difficult or impossible to identify, imaging modalities such as fluoroscopy or ultrasound can assist in proper identification of the intervertebral space and additionally provide information that improve safety during placement (fluoroscopy shows real-time needle advancement while ultrasound allows identification of the midline and measurement of the epidural space depth) [1, 2].

The spinal cord is continuous with the brainstem cranially and extends down to the L1 level in adults and L3 level in infants where it terminates as the conus medullaris. It is enveloped by a dural sac, which terminates more caudally, at the S2 level. The subarachnoid space between L1/L3 (adults and infants, respectively) and S2 is filled by the filum terminale internum (an extension of the pia mater) and spinal nerve roots originating from the conus medullaris (called cauda equina). The cauda equina harbors the L2–L5, S1–S5, and coccygeal nerve pairs, which innervate the lower limbs, pelvic organs, and perineum. After reaching the termination of the dural sac at the S2 level, the filum terminale internum continues as filum terminale externum to the back of the first segment of the coccyx.

Three distinct layers surrounding the spinal cord (from innermost to outermost) are called pia mater, arachnoid mater, and dura mater. The pia mater is highly vascularized and closely attaches to the spinal cord. Between the arachnoid mater and the pia mater, a small space called the intrathecal (subarachnoid) space is found which contains the cerebrospinal fluid (CSF). It is the target area for local anesthetic deposition in spinal anesthesia. The dura mater is attached to the arachnoid mater. Between them, a potential space exists, the subdural space. Outside of the dura mater, another small space called the epidural space is found which serves as the target area for deposition of local anesthetics in epidural anesthesia. The epidural space contains fat, lymphatics, and blood vessels (venous plexus). The arachnoid is a nonvascular membrane that offers the most resistance (90%) to drug migration.

The spinal cord itself has 31 segments (8 cervical, 12 thoracic, 5 lumbar, 5 sacral, 1 coccygeal) with their respective spinal nerves and is composed of the inner gray matter and the surrounding white matter.

During spinal anesthetic placement using a midline approach, multiple ligaments are traversed after entering the skin and subcutaneous tissue: the first ligament encountered is the supraspinous ligament, followed by the interspinous ligament. The following ligamentum flavum has a distinct feel that may alert the anesthesiologist to the proximity of the epidural space. After passage through the ligamentum flavum, the needle crosses the epidural space and pierces through the dura mater and the arachnoid mater into the subarachnoid space where CSF is encountered. When using a paramedian approach, the supraspinous and interspinous ligaments are bypassed [1, 2].

The spinal cord is supplied by a single anterior spinal artery that originates from the vertebral arteries and a pair of posterior spinal arteries that are formed by the inferior cerebellar arteries. These three arteries receive contributions from intercostal and lumbar arteries via segmental spinal arteries. The largest of those tributaries is the artery of Adamkiewicz (Arteria radicularis magna). In 75% of individuals, the artery of Adamkiewicz originates on the left side between T8 and L1. The anterior spinal artery supplies the anterior 2/3 of the spinal cord while the posterior spinal arteries cover the posterior 1/3.

Venous blood drains into three anterior and three posterior spinal veins that connect to the internal venous plexus located in the epidural space [1, 2].

Local anesthetics reversibly bind to voltage-gated sodium channels and block sodium conduction. By decreasing sodium currents, the initiation and propagation of action potentials is interrupted. Local anesthetics cause a “state-dependent” block: while sodium channels can be found in three states (resting-closed, activated-open, inactivated-closed), local anesthetics have the greatest affinity for channels in the activated-open state (less so for the inactivated-closed state).

Local anesthetics injected into the spinal space gain access to the spinal nerve roots and dorsal root ganglia.

Traditional teaching states that smaller, unmyelinated fibers are blocked first while larger, myelinated fibers take a longer time for complete blockade. As a result, onset of blockade follows a temporal progression, starting with preganglionic sympathetic conduction supplied by B-fibers, followed by cold temperature sensation conveyed by C-fibers, pinprick sensation transmitted via A-delta fibers, touch sensation through A-beta fibers and, finally, motor function provided by A-alpha fibers. Recovery follows in reverse order. The differences in sensitivity account for a phenomenon called differential sensory block: the peak block height of a given anesthetic may vary according to the sensory modality that is being tested. Typically, the highest block level (largest dermatomal spread) is attained for cold sensation, followed by sensation to pinprick (1–2 levels lower) and touch (3–4 levels lower). Sympathetic block height can reach 2–6 dermatomes above sensory block level while motor block height is found to be 2–3 dermatomes below sensory block level.

The term “differential sensory block” was introduced based on studies in the 1950s. Since then, multiple investigators have reported conflicting results. Some were able to reproduce a difference in block height extension for different modalities while others did not find this effect. These differences also vary between the different local anesthetics. Another layer of complexity is added when contrasting basic and clinical science results. Experimental studies on rat sciatic nerves, for example, have shown a higher susceptibility to lidocaine-induced blockade in A-gamma/delta/alpha and beta fibers compared to C-fibers [1, 2].

Upper abdominal surgery (cholecystectomy) and C-section require a T4–5 level corresponding to the nipple line (T4). While the uterus is located in the lower abdomen, a T4 level is necessary to blunt stimuli caused by traction on the peritoneum and uterine exteriorization. Lower abdominal surgery requires a T6–8 level, with T6/7 being at the level of the xiphoid process. Anesthesia extending to the level of the umbilicus, corresponding to the T10 dermatome, is adequate for TURP, hip surgery, and vaginal delivery. Thigh surgery and lower limb surgery are covered by a level extending at least to the inguinal ligament which corresponds to T12/L1, while foot surgery only requires a L2/3 level (except in cases that use a tourniquet); the area under the tourniquet corresponds to L1–L4 dermatomes. Taking temporal regression of the block into account, one should factor in a safety margin. An anesthetic level involving the S2–5 dermatomes adequately covers hemorrhoidectomy and perineal surgery.

In the cardiovascular system, spinal anesthesia causes a decrease in cardiac output and systemic vascular resistance. Blockade of sympathetic fibers inhibits vasoconstriction and leads to pooling of blood in the splanchnic system and the lower extremities. A decrease in venous return (~preload) results in a decrease in cardiac output. If local anesthetic spread reaches T1–T4 levels, blockade of the cardiac accelerator fibers occurs causing a decrease in heart rate, which further lowers cardiac output [3, 4].

In addition to those changes, three reflexes are associated with spinal anesthesia which can worsen bradycardia and even lead to cardiac arrest:

Overall, the direct effects of spinal anesthesia on the respiratory system are minimal. Even with mid- to high-thoracic levels of anesthesia, respiratory indices including minute ventilation, tidal volume, and mean inspiratory flow rate are unaffected. The same is true for pulmonary gas exchange. A small decrease in vital capacity and maximal expiratory pressure and flow has been observed and is attributed to weakness of the abdominal muscles [3, 4].

Spinal anesthesia has consistently been found to reduce hypnotic and sedative requirements through yet poorly understood mechanisms. Theories to explain this finding include de-afferentiation phenomenon (blockade of the spinal nerves reduces afferent input to the brain, making the reticular activating system more sensitive to sedative/hypnotic drugs), rostral spread of local anesthetics within the CSF (with direct blocking effects of local anesthetics on brain centers), and absorption of local anesthetics leading to increased systemic levels [3, 4].

Bladder and urethral sphincters are controlled via the sacral spinal nerves (S2–S4). Blockade of those segments abolishes the urge to void, creating the risk of bladder overdistention and postoperative urinary retention in the setting of long-acting local anesthetics. Bladder overdistention can cause pain and hypertension, and stimulation of vagal afferents can also lead to bradycardia and hypotension. Therefore, consideration should be given to placement of an indwelling urinary catheter or ultrasound bladder volume monitoring with as-needed in-and-out catheterization [3, 4].

Ablation of sympathetic stimuli to the gastrointestinal tract generated at T6–L1 levels results in hyperperistalsis from unopposed parasympathetic stimulation. Nausea and vomiting can be secondary to this effect (or, in other cases, related to hypotension and hypoperfusion) [1, 2].

In general, the body thermo-regulative capabilities are more impaired by general anesthesia than by spinal anesthesia; nevertheless, vasodilation from sympathetic blockade causes a re-distribution of heat from core tissues (head, trunk, internal organs) to skin, upper and lower extremities. Also, shivering thresholds are reduced in the blocked segments, thereby reducing the ability to generate heat [3, 4].

Block level in spinal anesthesia is determined by drug, patient, and procedure factors [1, 2].